nim-codex/codex/sales/reservations.nim

670 lines
22 KiB
Nim
Raw Normal View History

[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
## Nim-Codex
## Copyright (c) 2022 Status Research & Development GmbH
## Licensed under either of
## * Apache License, version 2.0, ([LICENSE-APACHE](LICENSE-APACHE))
## * MIT license ([LICENSE-MIT](LICENSE-MIT))
## at your option.
## This file may not be copied, modified, or distributed except according to
## those terms.
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
##
## +--------------------------------------+
## | RESERVATION |
## +----------------------------------------+ |--------------------------------------|
## | AVAILABILITY | | ReservationId | id | PK |
## |----------------------------------------| |--------------------------------------|
## | AvailabilityId | id | PK |<-||-------o<-| AvailabilityId | availabilityId | FK |
## |----------------------------------------| |--------------------------------------|
## | UInt256 | totalSize | | | UInt256 | size | |
## |----------------------------------------| |--------------------------------------|
## | UInt256 | freeSize | | | UInt256 | slotIndex | |
## |----------------------------------------| +--------------------------------------+
## | UInt256 | duration | |
## |----------------------------------------|
## | UInt256 | minPrice | |
## |----------------------------------------|
## | UInt256 | maxCollateral | |
## +----------------------------------------+
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
import pkg/upraises
push: {.upraises: [].}
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
feat[marketplace]: add slot queue pausing (#752) * add seen flag * Add MockSlotQueueItem and better prioritisation tests * Update seen priority, and include in SlotQueueItem.init * Re-add processed slots to queue Re-add processed slots to queue if the sale was ignored or errored * add pausing of queue - when processing slots in queue, pause queue if item was marked seen - if availability size is increased, trigger onAvailabilityAdded callback - in sales, on availability added, clear 'seen' flags, then unpause the queue - when items pushed to the queue, unpause the queue * remove unused NoMatchingAvailabilityError from slotqueue The slot queue should also have nothing to do with availabilities * when all availabilities are empty, pause the queue An empty availability is defined as size < DefaultBlockSize as this means even the smallest possible request could not be served. However, this is up for discussion. * remove availability from onAvailabilitiesEmptied callback * refactor onAvailabilityAdded and onAvailabilitiesEmptied onAvailabilityAdded and onAvailabilitiesEmptied are now only called from reservations.update (and eventually reservations.delete once implemented). - Add empty routine for Availability and Reservation - Add allEmpty routine for Availability and Reservation, which returns true when all all Availability or Reservation objects in the datastore are empty. * SlotQueue test support updates * Sales module test support updates * Reservations module tests for queue pausing * Sales module tests for queue pausing Includes tests for sales states cancelled, errored, ignored to ensure onCleanUp is called with correct parameters * SlotQueue module tests for queue pausing * fix existing sales test * PR feedback - indent `self.unpause` - update comment for `clearSeenFlags` * reprocessSlot in SaleErrored only when coming from downloading * remove pausing of queue when availabilities are "emptied" Queue pausing when all availiabilies are "emptied" is not necessary, given that the node would not be able to service slots once all its availabilities' freeSize are too small for the slots in the queue, and would then be paused anyway. Add test that asserts the queue is paused once the freeSpace of availabilities drops too low to fill slots in the queue. * Update clearing of seen flags The asyncheapqueue update overload would need to check index bounds and ultimately a different solution was found using the mitems iterator. * fix test request.id was different before updating request.ask.slots, and that id was used to set the state in mockmarket. * Change filled/cleanup future to nil, so no await is needed * add wait to allow items to be added to queue * do not unpause queue when seen items are pushed * re-add seen item back to queue once paused Previously, when a seen item was processed, it was first popped off the queue, then the queue was paused waiting to process that item once the queue was unpaused. Now, when a seen item is processed, it is popped off the queue, the queue is paused, then the item is re-added to the queue and the queue will wait until unpaused before it will continue popping items off the queue. If the item was not re-added to the queue, it would have been processed immediately once unpaused, however there may have been other items with higher priority pushed to the queue in the meantime. The queue would not be unpaused if those added items were already seen. In particular, this may happen when ignored items due to lack of availability are re-added to a paused queue. Those ignored items will likely have a higher priority than the item that was just seen (due to it having been processed first), causing the queue to the be paused. * address PR comments
2024-05-26 00:38:38 +00:00
import std/sequtils
import std/sugar
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
import std/typetraits
import std/sequtils
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
import pkg/chronos
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
import pkg/datastore
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
import pkg/nimcrypto
import pkg/questionable
import pkg/questionable/results
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
import pkg/stint
import pkg/stew/byteutils
feat[marketplace]: add slot queue pausing (#752) * add seen flag * Add MockSlotQueueItem and better prioritisation tests * Update seen priority, and include in SlotQueueItem.init * Re-add processed slots to queue Re-add processed slots to queue if the sale was ignored or errored * add pausing of queue - when processing slots in queue, pause queue if item was marked seen - if availability size is increased, trigger onAvailabilityAdded callback - in sales, on availability added, clear 'seen' flags, then unpause the queue - when items pushed to the queue, unpause the queue * remove unused NoMatchingAvailabilityError from slotqueue The slot queue should also have nothing to do with availabilities * when all availabilities are empty, pause the queue An empty availability is defined as size < DefaultBlockSize as this means even the smallest possible request could not be served. However, this is up for discussion. * remove availability from onAvailabilitiesEmptied callback * refactor onAvailabilityAdded and onAvailabilitiesEmptied onAvailabilityAdded and onAvailabilitiesEmptied are now only called from reservations.update (and eventually reservations.delete once implemented). - Add empty routine for Availability and Reservation - Add allEmpty routine for Availability and Reservation, which returns true when all all Availability or Reservation objects in the datastore are empty. * SlotQueue test support updates * Sales module test support updates * Reservations module tests for queue pausing * Sales module tests for queue pausing Includes tests for sales states cancelled, errored, ignored to ensure onCleanUp is called with correct parameters * SlotQueue module tests for queue pausing * fix existing sales test * PR feedback - indent `self.unpause` - update comment for `clearSeenFlags` * reprocessSlot in SaleErrored only when coming from downloading * remove pausing of queue when availabilities are "emptied" Queue pausing when all availiabilies are "emptied" is not necessary, given that the node would not be able to service slots once all its availabilities' freeSize are too small for the slots in the queue, and would then be paused anyway. Add test that asserts the queue is paused once the freeSpace of availabilities drops too low to fill slots in the queue. * Update clearing of seen flags The asyncheapqueue update overload would need to check index bounds and ultimately a different solution was found using the mitems iterator. * fix test request.id was different before updating request.ask.slots, and that id was used to set the state in mockmarket. * Change filled/cleanup future to nil, so no await is needed * add wait to allow items to be added to queue * do not unpause queue when seen items are pushed * re-add seen item back to queue once paused Previously, when a seen item was processed, it was first popped off the queue, then the queue was paused waiting to process that item once the queue was unpaused. Now, when a seen item is processed, it is popped off the queue, the queue is paused, then the item is re-added to the queue and the queue will wait until unpaused before it will continue popping items off the queue. If the item was not re-added to the queue, it would have been processed immediately once unpaused, however there may have been other items with higher priority pushed to the queue in the meantime. The queue would not be unpaused if those added items were already seen. In particular, this may happen when ignored items due to lack of availability are re-added to a paused queue. Those ignored items will likely have a higher priority than the item that was just seen (due to it having been processed first), causing the queue to the be paused. * address PR comments
2024-05-26 00:38:38 +00:00
import ../codextypes
feat: create logging proxy (#663) * implement a logging proxy The logging proxy: - prevents the need to import chronicles (as well as export except toJson), - prevents the need to override `writeValue` or use or import nim-json-seralization elsewhere in the codebase, allowing for sole use of utils/json for de/serialization, - and handles json formatting correctly in chronicles json sinks * Rename logging -> logutils to avoid ambiguity with common names * clean up * add setProperty for JsonRecord, remove nim-json-serialization conflict * Allow specifying textlines and json format separately Not specifying a LogFormat will apply the formatting to both textlines and json sinks. Specifying a LogFormat will apply the formatting to only that sink. * remove unneeded usages of std/json We only need to import utils/json instead of std/json * move serialization from rest/json to utils/json so it can be shared * fix NoColors ambiguity Was causing unit tests to fail on Windows. * Remove nre usage to fix Windows error Windows was erroring with `could not load: pcre64.dll`. Instead of fixing that error, remove the pcre usage :) * Add logutils module doc * Shorten logutils.formatIt for `NBytes` Both json and textlines formatIt were not needed, and could be combined into one formatIt * remove debug integration test config debug output and logformat of json for integration test logs * Use ## module doc to support docgen * bump nim-poseidon2 to export fromBytes Before the changes in this branch, fromBytes was likely being resolved by nim-stew, or other dependency. With the changes in this branch, that dependency was removed and fromBytes could no longer be resolved. By exporting fromBytes from nim-poseidon, the correct resolution is now happening. * fixes to get compiling after rebasing master * Add support for Result types being logged using formatIt
2024-01-23 07:35:03 +00:00
import ../logutils
import ../clock
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
import ../stores
import ../market
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
import ../contracts/requests
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
import ../utils/json
import ../units
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
export requests
feat: create logging proxy (#663) * implement a logging proxy The logging proxy: - prevents the need to import chronicles (as well as export except toJson), - prevents the need to override `writeValue` or use or import nim-json-seralization elsewhere in the codebase, allowing for sole use of utils/json for de/serialization, - and handles json formatting correctly in chronicles json sinks * Rename logging -> logutils to avoid ambiguity with common names * clean up * add setProperty for JsonRecord, remove nim-json-serialization conflict * Allow specifying textlines and json format separately Not specifying a LogFormat will apply the formatting to both textlines and json sinks. Specifying a LogFormat will apply the formatting to only that sink. * remove unneeded usages of std/json We only need to import utils/json instead of std/json * move serialization from rest/json to utils/json so it can be shared * fix NoColors ambiguity Was causing unit tests to fail on Windows. * Remove nre usage to fix Windows error Windows was erroring with `could not load: pcre64.dll`. Instead of fixing that error, remove the pcre usage :) * Add logutils module doc * Shorten logutils.formatIt for `NBytes` Both json and textlines formatIt were not needed, and could be combined into one formatIt * remove debug integration test config debug output and logformat of json for integration test logs * Use ## module doc to support docgen * bump nim-poseidon2 to export fromBytes Before the changes in this branch, fromBytes was likely being resolved by nim-stew, or other dependency. With the changes in this branch, that dependency was removed and fromBytes could no longer be resolved. By exporting fromBytes from nim-poseidon, the correct resolution is now happening. * fixes to get compiling after rebasing master * Add support for Result types being logged using formatIt
2024-01-23 07:35:03 +00:00
export logutils
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
logScope:
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
topics = "sales reservations"
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
type
AvailabilityId* = distinct array[32, byte]
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
ReservationId* = distinct array[32, byte]
SomeStorableObject = Availability | Reservation
SomeStorableId = AvailabilityId | ReservationId
Availability* = ref object
id* {.serialize.}: AvailabilityId
totalSize* {.serialize.}: UInt256
freeSize* {.serialize.}: UInt256
duration* {.serialize.}: UInt256
minPrice* {.serialize.}: UInt256 # minimal price paid for the whole hosted slot for the request's duration
maxCollateral* {.serialize.}: UInt256
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
Reservation* = ref object
id* {.serialize.}: ReservationId
availabilityId* {.serialize.}: AvailabilityId
size* {.serialize.}: UInt256
requestId* {.serialize.}: RequestId
slotIndex* {.serialize.}: UInt256
Reservations* = ref object of RootObj
availabilityLock: AsyncLock # Lock for protecting assertions of availability's sizes when searching for matching availability
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
repo: RepoStore
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
onAvailabilityAdded: ?OnAvailabilityAdded
GetNext* = proc(): Future[?seq[byte]] {.upraises: [], gcsafe, closure.}
IterDispose* = proc(): Future[?!void] {.gcsafe, closure.}
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
OnAvailabilityAdded* = proc(availability: Availability): Future[void] {.upraises: [], gcsafe.}
StorableIter* = ref object
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
finished*: bool
next*: GetNext
dispose*: IterDispose
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
ReservationsError* = object of CodexError
ReserveFailedError* = object of ReservationsError
ReleaseFailedError* = object of ReservationsError
DeleteFailedError* = object of ReservationsError
GetFailedError* = object of ReservationsError
NotExistsError* = object of ReservationsError
SerializationError* = object of ReservationsError
UpdateFailedError* = object of ReservationsError
BytesOutOfBoundsError* = object of ReservationsError
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
const
SalesKey = (CodexMetaKey / "sales").tryGet # TODO: move to sales module
ReservationsKey = (SalesKey / "reservations").tryGet
proc hash*(x: AvailabilityId): Hash {.borrow.}
feat[marketplace]: add slot queue pausing (#752) * add seen flag * Add MockSlotQueueItem and better prioritisation tests * Update seen priority, and include in SlotQueueItem.init * Re-add processed slots to queue Re-add processed slots to queue if the sale was ignored or errored * add pausing of queue - when processing slots in queue, pause queue if item was marked seen - if availability size is increased, trigger onAvailabilityAdded callback - in sales, on availability added, clear 'seen' flags, then unpause the queue - when items pushed to the queue, unpause the queue * remove unused NoMatchingAvailabilityError from slotqueue The slot queue should also have nothing to do with availabilities * when all availabilities are empty, pause the queue An empty availability is defined as size < DefaultBlockSize as this means even the smallest possible request could not be served. However, this is up for discussion. * remove availability from onAvailabilitiesEmptied callback * refactor onAvailabilityAdded and onAvailabilitiesEmptied onAvailabilityAdded and onAvailabilitiesEmptied are now only called from reservations.update (and eventually reservations.delete once implemented). - Add empty routine for Availability and Reservation - Add allEmpty routine for Availability and Reservation, which returns true when all all Availability or Reservation objects in the datastore are empty. * SlotQueue test support updates * Sales module test support updates * Reservations module tests for queue pausing * Sales module tests for queue pausing Includes tests for sales states cancelled, errored, ignored to ensure onCleanUp is called with correct parameters * SlotQueue module tests for queue pausing * fix existing sales test * PR feedback - indent `self.unpause` - update comment for `clearSeenFlags` * reprocessSlot in SaleErrored only when coming from downloading * remove pausing of queue when availabilities are "emptied" Queue pausing when all availiabilies are "emptied" is not necessary, given that the node would not be able to service slots once all its availabilities' freeSize are too small for the slots in the queue, and would then be paused anyway. Add test that asserts the queue is paused once the freeSpace of availabilities drops too low to fill slots in the queue. * Update clearing of seen flags The asyncheapqueue update overload would need to check index bounds and ultimately a different solution was found using the mitems iterator. * fix test request.id was different before updating request.ask.slots, and that id was used to set the state in mockmarket. * Change filled/cleanup future to nil, so no await is needed * add wait to allow items to be added to queue * do not unpause queue when seen items are pushed * re-add seen item back to queue once paused Previously, when a seen item was processed, it was first popped off the queue, then the queue was paused waiting to process that item once the queue was unpaused. Now, when a seen item is processed, it is popped off the queue, the queue is paused, then the item is re-added to the queue and the queue will wait until unpaused before it will continue popping items off the queue. If the item was not re-added to the queue, it would have been processed immediately once unpaused, however there may have been other items with higher priority pushed to the queue in the meantime. The queue would not be unpaused if those added items were already seen. In particular, this may happen when ignored items due to lack of availability are re-added to a paused queue. Those ignored items will likely have a higher priority than the item that was just seen (due to it having been processed first), causing the queue to the be paused. * address PR comments
2024-05-26 00:38:38 +00:00
proc all*(self: Reservations, T: type SomeStorableObject): Future[?!seq[T]] {.async.}
template withLock(lock, body) =
try:
await lock.acquire()
body
finally:
if lock.locked:
lock.release()
proc new*(T: type Reservations,
repo: RepoStore): Reservations =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
T(availabilityLock: newAsyncLock(),repo: repo)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
proc init*(
_: type Availability,
totalSize: UInt256,
freeSize: UInt256,
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
duration: UInt256,
minPrice: UInt256,
maxCollateral: UInt256): Availability =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
var id: array[32, byte]
doAssert randomBytes(id) == 32
Availability(id: AvailabilityId(id), totalSize:totalSize, freeSize: freeSize, duration: duration, minPrice: minPrice, maxCollateral: maxCollateral)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc init*(
_: type Reservation,
availabilityId: AvailabilityId,
size: UInt256,
requestId: RequestId,
slotIndex: UInt256
): Reservation =
var id: array[32, byte]
doAssert randomBytes(id) == 32
Reservation(id: ReservationId(id), availabilityId: availabilityId, size: size, requestId: requestId, slotIndex: slotIndex)
func toArray(id: SomeStorableId): array[32, byte] =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
array[32, byte](id)
proc `==`*(x, y: AvailabilityId): bool {.borrow.}
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc `==`*(x, y: ReservationId): bool {.borrow.}
proc `==`*(x, y: Reservation): bool =
x.id == y.id
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
proc `==`*(x, y: Availability): bool =
x.id == y.id
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc `$`*(id: SomeStorableId): string = id.toArray.toHex
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc toErr[E1: ref CatchableError, E2: ReservationsError](
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
e1: E1,
_: type E2,
msg: string = e1.msg): ref E2 =
return newException(E2, msg, e1)
feat: create logging proxy (#663) * implement a logging proxy The logging proxy: - prevents the need to import chronicles (as well as export except toJson), - prevents the need to override `writeValue` or use or import nim-json-seralization elsewhere in the codebase, allowing for sole use of utils/json for de/serialization, - and handles json formatting correctly in chronicles json sinks * Rename logging -> logutils to avoid ambiguity with common names * clean up * add setProperty for JsonRecord, remove nim-json-serialization conflict * Allow specifying textlines and json format separately Not specifying a LogFormat will apply the formatting to both textlines and json sinks. Specifying a LogFormat will apply the formatting to only that sink. * remove unneeded usages of std/json We only need to import utils/json instead of std/json * move serialization from rest/json to utils/json so it can be shared * fix NoColors ambiguity Was causing unit tests to fail on Windows. * Remove nre usage to fix Windows error Windows was erroring with `could not load: pcre64.dll`. Instead of fixing that error, remove the pcre usage :) * Add logutils module doc * Shorten logutils.formatIt for `NBytes` Both json and textlines formatIt were not needed, and could be combined into one formatIt * remove debug integration test config debug output and logformat of json for integration test logs * Use ## module doc to support docgen * bump nim-poseidon2 to export fromBytes Before the changes in this branch, fromBytes was likely being resolved by nim-stew, or other dependency. With the changes in this branch, that dependency was removed and fromBytes could no longer be resolved. By exporting fromBytes from nim-poseidon, the correct resolution is now happening. * fixes to get compiling after rebasing master * Add support for Result types being logged using formatIt
2024-01-23 07:35:03 +00:00
logutils.formatIt(LogFormat.textLines, SomeStorableId): it.short0xHexLog
logutils.formatIt(LogFormat.json, SomeStorableId): it.to0xHexLog
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc `onAvailabilityAdded=`*(self: Reservations,
onAvailabilityAdded: OnAvailabilityAdded) =
self.onAvailabilityAdded = some onAvailabilityAdded
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
func key*(id: AvailabilityId): ?!Key =
## sales / reservations / <availabilityId>
(ReservationsKey / $id)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
func key*(reservationId: ReservationId, availabilityId: AvailabilityId): ?!Key =
## sales / reservations / <availabilityId> / <reservationId>
(availabilityId.key / $reservationId)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
func key*(availability: Availability): ?!Key =
return availability.id.key
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
func key*(reservation: Reservation): ?!Key =
return key(reservation.id, reservation.availabilityId)
func available*(self: Reservations): uint = self.repo.available.uint
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
func hasAvailable*(self: Reservations, bytes: uint): bool =
self.repo.available(bytes.NBytes)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
proc exists*(
self: Reservations,
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
key: Key): Future[bool] {.async.} =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
let exists = await self.repo.metaDs.ds.contains(key)
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return exists
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc getImpl(
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
self: Reservations,
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
key: Key): Future[?!seq[byte]] {.async.} =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
if not await self.exists(key):
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
let err = newException(NotExistsError, "object with key " & $key & " does not exist")
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
return failure(err)
without serialized =? await self.repo.metaDs.ds.get(key), error:
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return failure(error.toErr(GetFailedError))
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return success serialized
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc get*(
self: Reservations,
key: Key,
T: type SomeStorableObject): Future[?!T] {.async.} =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
without serialized =? await self.getImpl(key), error:
return failure(error)
without obj =? T.fromJson(serialized), error:
return failure(error.toErr(SerializationError))
return success obj
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
proc updateImpl(
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
self: Reservations,
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
obj: SomeStorableObject): Future[?!void] {.async.} =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
trace "updating " & $(obj.type), id = obj.id
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
without key =? obj.key, error:
return failure(error)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
if err =? (await self.repo.metaDs.ds.put(
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
key,
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
@(obj.toJson.toBytes)
)).errorOption:
return failure(err.toErr(UpdateFailedError))
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
return success()
proc updateAvailability(
self: Reservations,
obj: Availability): Future[?!void] {.async.} =
logScope:
availabilityId = obj.id
without key =? obj.key, error:
return failure(error)
feat[marketplace]: add slot queue pausing (#752) * add seen flag * Add MockSlotQueueItem and better prioritisation tests * Update seen priority, and include in SlotQueueItem.init * Re-add processed slots to queue Re-add processed slots to queue if the sale was ignored or errored * add pausing of queue - when processing slots in queue, pause queue if item was marked seen - if availability size is increased, trigger onAvailabilityAdded callback - in sales, on availability added, clear 'seen' flags, then unpause the queue - when items pushed to the queue, unpause the queue * remove unused NoMatchingAvailabilityError from slotqueue The slot queue should also have nothing to do with availabilities * when all availabilities are empty, pause the queue An empty availability is defined as size < DefaultBlockSize as this means even the smallest possible request could not be served. However, this is up for discussion. * remove availability from onAvailabilitiesEmptied callback * refactor onAvailabilityAdded and onAvailabilitiesEmptied onAvailabilityAdded and onAvailabilitiesEmptied are now only called from reservations.update (and eventually reservations.delete once implemented). - Add empty routine for Availability and Reservation - Add allEmpty routine for Availability and Reservation, which returns true when all all Availability or Reservation objects in the datastore are empty. * SlotQueue test support updates * Sales module test support updates * Reservations module tests for queue pausing * Sales module tests for queue pausing Includes tests for sales states cancelled, errored, ignored to ensure onCleanUp is called with correct parameters * SlotQueue module tests for queue pausing * fix existing sales test * PR feedback - indent `self.unpause` - update comment for `clearSeenFlags` * reprocessSlot in SaleErrored only when coming from downloading * remove pausing of queue when availabilities are "emptied" Queue pausing when all availiabilies are "emptied" is not necessary, given that the node would not be able to service slots once all its availabilities' freeSize are too small for the slots in the queue, and would then be paused anyway. Add test that asserts the queue is paused once the freeSpace of availabilities drops too low to fill slots in the queue. * Update clearing of seen flags The asyncheapqueue update overload would need to check index bounds and ultimately a different solution was found using the mitems iterator. * fix test request.id was different before updating request.ask.slots, and that id was used to set the state in mockmarket. * Change filled/cleanup future to nil, so no await is needed * add wait to allow items to be added to queue * do not unpause queue when seen items are pushed * re-add seen item back to queue once paused Previously, when a seen item was processed, it was first popped off the queue, then the queue was paused waiting to process that item once the queue was unpaused. Now, when a seen item is processed, it is popped off the queue, the queue is paused, then the item is re-added to the queue and the queue will wait until unpaused before it will continue popping items off the queue. If the item was not re-added to the queue, it would have been processed immediately once unpaused, however there may have been other items with higher priority pushed to the queue in the meantime. The queue would not be unpaused if those added items were already seen. In particular, this may happen when ignored items due to lack of availability are re-added to a paused queue. Those ignored items will likely have a higher priority than the item that was just seen (due to it having been processed first), causing the queue to the be paused. * address PR comments
2024-05-26 00:38:38 +00:00
without oldAvailability =? await self.get(key, Availability), err:
if err of NotExistsError:
trace "Creating new Availability"
feat[marketplace]: add slot queue pausing (#752) * add seen flag * Add MockSlotQueueItem and better prioritisation tests * Update seen priority, and include in SlotQueueItem.init * Re-add processed slots to queue Re-add processed slots to queue if the sale was ignored or errored * add pausing of queue - when processing slots in queue, pause queue if item was marked seen - if availability size is increased, trigger onAvailabilityAdded callback - in sales, on availability added, clear 'seen' flags, then unpause the queue - when items pushed to the queue, unpause the queue * remove unused NoMatchingAvailabilityError from slotqueue The slot queue should also have nothing to do with availabilities * when all availabilities are empty, pause the queue An empty availability is defined as size < DefaultBlockSize as this means even the smallest possible request could not be served. However, this is up for discussion. * remove availability from onAvailabilitiesEmptied callback * refactor onAvailabilityAdded and onAvailabilitiesEmptied onAvailabilityAdded and onAvailabilitiesEmptied are now only called from reservations.update (and eventually reservations.delete once implemented). - Add empty routine for Availability and Reservation - Add allEmpty routine for Availability and Reservation, which returns true when all all Availability or Reservation objects in the datastore are empty. * SlotQueue test support updates * Sales module test support updates * Reservations module tests for queue pausing * Sales module tests for queue pausing Includes tests for sales states cancelled, errored, ignored to ensure onCleanUp is called with correct parameters * SlotQueue module tests for queue pausing * fix existing sales test * PR feedback - indent `self.unpause` - update comment for `clearSeenFlags` * reprocessSlot in SaleErrored only when coming from downloading * remove pausing of queue when availabilities are "emptied" Queue pausing when all availiabilies are "emptied" is not necessary, given that the node would not be able to service slots once all its availabilities' freeSize are too small for the slots in the queue, and would then be paused anyway. Add test that asserts the queue is paused once the freeSpace of availabilities drops too low to fill slots in the queue. * Update clearing of seen flags The asyncheapqueue update overload would need to check index bounds and ultimately a different solution was found using the mitems iterator. * fix test request.id was different before updating request.ask.slots, and that id was used to set the state in mockmarket. * Change filled/cleanup future to nil, so no await is needed * add wait to allow items to be added to queue * do not unpause queue when seen items are pushed * re-add seen item back to queue once paused Previously, when a seen item was processed, it was first popped off the queue, then the queue was paused waiting to process that item once the queue was unpaused. Now, when a seen item is processed, it is popped off the queue, the queue is paused, then the item is re-added to the queue and the queue will wait until unpaused before it will continue popping items off the queue. If the item was not re-added to the queue, it would have been processed immediately once unpaused, however there may have been other items with higher priority pushed to the queue in the meantime. The queue would not be unpaused if those added items were already seen. In particular, this may happen when ignored items due to lack of availability are re-added to a paused queue. Those ignored items will likely have a higher priority than the item that was just seen (due to it having been processed first), causing the queue to the be paused. * address PR comments
2024-05-26 00:38:38 +00:00
let res = await self.updateImpl(obj)
# inform subscribers that Availability has been added
if onAvailabilityAdded =? self.onAvailabilityAdded:
# when chronos v4 is implemented, and OnAvailabilityAdded is annotated
# with async:(raises:[]), we can remove this try/catch as we know, with
# certainty, that nothing will be raised
try:
await onAvailabilityAdded(obj)
except CancelledError as e:
raise e
except CatchableError as e:
# we don't have any insight into types of exceptions that
# `onAvailabilityAdded` can raise because it is caller-defined
warn "Unknown error during 'onAvailabilityAdded' callback", error = e.msg
feat[marketplace]: add slot queue pausing (#752) * add seen flag * Add MockSlotQueueItem and better prioritisation tests * Update seen priority, and include in SlotQueueItem.init * Re-add processed slots to queue Re-add processed slots to queue if the sale was ignored or errored * add pausing of queue - when processing slots in queue, pause queue if item was marked seen - if availability size is increased, trigger onAvailabilityAdded callback - in sales, on availability added, clear 'seen' flags, then unpause the queue - when items pushed to the queue, unpause the queue * remove unused NoMatchingAvailabilityError from slotqueue The slot queue should also have nothing to do with availabilities * when all availabilities are empty, pause the queue An empty availability is defined as size < DefaultBlockSize as this means even the smallest possible request could not be served. However, this is up for discussion. * remove availability from onAvailabilitiesEmptied callback * refactor onAvailabilityAdded and onAvailabilitiesEmptied onAvailabilityAdded and onAvailabilitiesEmptied are now only called from reservations.update (and eventually reservations.delete once implemented). - Add empty routine for Availability and Reservation - Add allEmpty routine for Availability and Reservation, which returns true when all all Availability or Reservation objects in the datastore are empty. * SlotQueue test support updates * Sales module test support updates * Reservations module tests for queue pausing * Sales module tests for queue pausing Includes tests for sales states cancelled, errored, ignored to ensure onCleanUp is called with correct parameters * SlotQueue module tests for queue pausing * fix existing sales test * PR feedback - indent `self.unpause` - update comment for `clearSeenFlags` * reprocessSlot in SaleErrored only when coming from downloading * remove pausing of queue when availabilities are "emptied" Queue pausing when all availiabilies are "emptied" is not necessary, given that the node would not be able to service slots once all its availabilities' freeSize are too small for the slots in the queue, and would then be paused anyway. Add test that asserts the queue is paused once the freeSpace of availabilities drops too low to fill slots in the queue. * Update clearing of seen flags The asyncheapqueue update overload would need to check index bounds and ultimately a different solution was found using the mitems iterator. * fix test request.id was different before updating request.ask.slots, and that id was used to set the state in mockmarket. * Change filled/cleanup future to nil, so no await is needed * add wait to allow items to be added to queue * do not unpause queue when seen items are pushed * re-add seen item back to queue once paused Previously, when a seen item was processed, it was first popped off the queue, then the queue was paused waiting to process that item once the queue was unpaused. Now, when a seen item is processed, it is popped off the queue, the queue is paused, then the item is re-added to the queue and the queue will wait until unpaused before it will continue popping items off the queue. If the item was not re-added to the queue, it would have been processed immediately once unpaused, however there may have been other items with higher priority pushed to the queue in the meantime. The queue would not be unpaused if those added items were already seen. In particular, this may happen when ignored items due to lack of availability are re-added to a paused queue. Those ignored items will likely have a higher priority than the item that was just seen (due to it having been processed first), causing the queue to the be paused. * address PR comments
2024-05-26 00:38:38 +00:00
return res
else:
return failure(err)
feat[marketplace]: add slot queue pausing (#752) * add seen flag * Add MockSlotQueueItem and better prioritisation tests * Update seen priority, and include in SlotQueueItem.init * Re-add processed slots to queue Re-add processed slots to queue if the sale was ignored or errored * add pausing of queue - when processing slots in queue, pause queue if item was marked seen - if availability size is increased, trigger onAvailabilityAdded callback - in sales, on availability added, clear 'seen' flags, then unpause the queue - when items pushed to the queue, unpause the queue * remove unused NoMatchingAvailabilityError from slotqueue The slot queue should also have nothing to do with availabilities * when all availabilities are empty, pause the queue An empty availability is defined as size < DefaultBlockSize as this means even the smallest possible request could not be served. However, this is up for discussion. * remove availability from onAvailabilitiesEmptied callback * refactor onAvailabilityAdded and onAvailabilitiesEmptied onAvailabilityAdded and onAvailabilitiesEmptied are now only called from reservations.update (and eventually reservations.delete once implemented). - Add empty routine for Availability and Reservation - Add allEmpty routine for Availability and Reservation, which returns true when all all Availability or Reservation objects in the datastore are empty. * SlotQueue test support updates * Sales module test support updates * Reservations module tests for queue pausing * Sales module tests for queue pausing Includes tests for sales states cancelled, errored, ignored to ensure onCleanUp is called with correct parameters * SlotQueue module tests for queue pausing * fix existing sales test * PR feedback - indent `self.unpause` - update comment for `clearSeenFlags` * reprocessSlot in SaleErrored only when coming from downloading * remove pausing of queue when availabilities are "emptied" Queue pausing when all availiabilies are "emptied" is not necessary, given that the node would not be able to service slots once all its availabilities' freeSize are too small for the slots in the queue, and would then be paused anyway. Add test that asserts the queue is paused once the freeSpace of availabilities drops too low to fill slots in the queue. * Update clearing of seen flags The asyncheapqueue update overload would need to check index bounds and ultimately a different solution was found using the mitems iterator. * fix test request.id was different before updating request.ask.slots, and that id was used to set the state in mockmarket. * Change filled/cleanup future to nil, so no await is needed * add wait to allow items to be added to queue * do not unpause queue when seen items are pushed * re-add seen item back to queue once paused Previously, when a seen item was processed, it was first popped off the queue, then the queue was paused waiting to process that item once the queue was unpaused. Now, when a seen item is processed, it is popped off the queue, the queue is paused, then the item is re-added to the queue and the queue will wait until unpaused before it will continue popping items off the queue. If the item was not re-added to the queue, it would have been processed immediately once unpaused, however there may have been other items with higher priority pushed to the queue in the meantime. The queue would not be unpaused if those added items were already seen. In particular, this may happen when ignored items due to lack of availability are re-added to a paused queue. Those ignored items will likely have a higher priority than the item that was just seen (due to it having been processed first), causing the queue to the be paused. * address PR comments
2024-05-26 00:38:38 +00:00
# Sizing of the availability changed, we need to adjust the repo reservation accordingly
if oldAvailability.totalSize != obj.totalSize:
trace "totalSize changed, updating repo reservation"
feat[marketplace]: add slot queue pausing (#752) * add seen flag * Add MockSlotQueueItem and better prioritisation tests * Update seen priority, and include in SlotQueueItem.init * Re-add processed slots to queue Re-add processed slots to queue if the sale was ignored or errored * add pausing of queue - when processing slots in queue, pause queue if item was marked seen - if availability size is increased, trigger onAvailabilityAdded callback - in sales, on availability added, clear 'seen' flags, then unpause the queue - when items pushed to the queue, unpause the queue * remove unused NoMatchingAvailabilityError from slotqueue The slot queue should also have nothing to do with availabilities * when all availabilities are empty, pause the queue An empty availability is defined as size < DefaultBlockSize as this means even the smallest possible request could not be served. However, this is up for discussion. * remove availability from onAvailabilitiesEmptied callback * refactor onAvailabilityAdded and onAvailabilitiesEmptied onAvailabilityAdded and onAvailabilitiesEmptied are now only called from reservations.update (and eventually reservations.delete once implemented). - Add empty routine for Availability and Reservation - Add allEmpty routine for Availability and Reservation, which returns true when all all Availability or Reservation objects in the datastore are empty. * SlotQueue test support updates * Sales module test support updates * Reservations module tests for queue pausing * Sales module tests for queue pausing Includes tests for sales states cancelled, errored, ignored to ensure onCleanUp is called with correct parameters * SlotQueue module tests for queue pausing * fix existing sales test * PR feedback - indent `self.unpause` - update comment for `clearSeenFlags` * reprocessSlot in SaleErrored only when coming from downloading * remove pausing of queue when availabilities are "emptied" Queue pausing when all availiabilies are "emptied" is not necessary, given that the node would not be able to service slots once all its availabilities' freeSize are too small for the slots in the queue, and would then be paused anyway. Add test that asserts the queue is paused once the freeSpace of availabilities drops too low to fill slots in the queue. * Update clearing of seen flags The asyncheapqueue update overload would need to check index bounds and ultimately a different solution was found using the mitems iterator. * fix test request.id was different before updating request.ask.slots, and that id was used to set the state in mockmarket. * Change filled/cleanup future to nil, so no await is needed * add wait to allow items to be added to queue * do not unpause queue when seen items are pushed * re-add seen item back to queue once paused Previously, when a seen item was processed, it was first popped off the queue, then the queue was paused waiting to process that item once the queue was unpaused. Now, when a seen item is processed, it is popped off the queue, the queue is paused, then the item is re-added to the queue and the queue will wait until unpaused before it will continue popping items off the queue. If the item was not re-added to the queue, it would have been processed immediately once unpaused, however there may have been other items with higher priority pushed to the queue in the meantime. The queue would not be unpaused if those added items were already seen. In particular, this may happen when ignored items due to lack of availability are re-added to a paused queue. Those ignored items will likely have a higher priority than the item that was just seen (due to it having been processed first), causing the queue to the be paused. * address PR comments
2024-05-26 00:38:38 +00:00
if oldAvailability.totalSize < obj.totalSize: # storage added
if reserveErr =? (await self.repo.reserve((obj.totalSize - oldAvailability.totalSize).truncate(uint).NBytes)).errorOption:
feat[marketplace]: add slot queue pausing (#752) * add seen flag * Add MockSlotQueueItem and better prioritisation tests * Update seen priority, and include in SlotQueueItem.init * Re-add processed slots to queue Re-add processed slots to queue if the sale was ignored or errored * add pausing of queue - when processing slots in queue, pause queue if item was marked seen - if availability size is increased, trigger onAvailabilityAdded callback - in sales, on availability added, clear 'seen' flags, then unpause the queue - when items pushed to the queue, unpause the queue * remove unused NoMatchingAvailabilityError from slotqueue The slot queue should also have nothing to do with availabilities * when all availabilities are empty, pause the queue An empty availability is defined as size < DefaultBlockSize as this means even the smallest possible request could not be served. However, this is up for discussion. * remove availability from onAvailabilitiesEmptied callback * refactor onAvailabilityAdded and onAvailabilitiesEmptied onAvailabilityAdded and onAvailabilitiesEmptied are now only called from reservations.update (and eventually reservations.delete once implemented). - Add empty routine for Availability and Reservation - Add allEmpty routine for Availability and Reservation, which returns true when all all Availability or Reservation objects in the datastore are empty. * SlotQueue test support updates * Sales module test support updates * Reservations module tests for queue pausing * Sales module tests for queue pausing Includes tests for sales states cancelled, errored, ignored to ensure onCleanUp is called with correct parameters * SlotQueue module tests for queue pausing * fix existing sales test * PR feedback - indent `self.unpause` - update comment for `clearSeenFlags` * reprocessSlot in SaleErrored only when coming from downloading * remove pausing of queue when availabilities are "emptied" Queue pausing when all availiabilies are "emptied" is not necessary, given that the node would not be able to service slots once all its availabilities' freeSize are too small for the slots in the queue, and would then be paused anyway. Add test that asserts the queue is paused once the freeSpace of availabilities drops too low to fill slots in the queue. * Update clearing of seen flags The asyncheapqueue update overload would need to check index bounds and ultimately a different solution was found using the mitems iterator. * fix test request.id was different before updating request.ask.slots, and that id was used to set the state in mockmarket. * Change filled/cleanup future to nil, so no await is needed * add wait to allow items to be added to queue * do not unpause queue when seen items are pushed * re-add seen item back to queue once paused Previously, when a seen item was processed, it was first popped off the queue, then the queue was paused waiting to process that item once the queue was unpaused. Now, when a seen item is processed, it is popped off the queue, the queue is paused, then the item is re-added to the queue and the queue will wait until unpaused before it will continue popping items off the queue. If the item was not re-added to the queue, it would have been processed immediately once unpaused, however there may have been other items with higher priority pushed to the queue in the meantime. The queue would not be unpaused if those added items were already seen. In particular, this may happen when ignored items due to lack of availability are re-added to a paused queue. Those ignored items will likely have a higher priority than the item that was just seen (due to it having been processed first), causing the queue to the be paused. * address PR comments
2024-05-26 00:38:38 +00:00
return failure(reserveErr.toErr(ReserveFailedError))
elif oldAvailability.totalSize > obj.totalSize: # storage removed
if reserveErr =? (await self.repo.release((oldAvailability.totalSize - obj.totalSize).truncate(uint).NBytes)).errorOption:
feat[marketplace]: add slot queue pausing (#752) * add seen flag * Add MockSlotQueueItem and better prioritisation tests * Update seen priority, and include in SlotQueueItem.init * Re-add processed slots to queue Re-add processed slots to queue if the sale was ignored or errored * add pausing of queue - when processing slots in queue, pause queue if item was marked seen - if availability size is increased, trigger onAvailabilityAdded callback - in sales, on availability added, clear 'seen' flags, then unpause the queue - when items pushed to the queue, unpause the queue * remove unused NoMatchingAvailabilityError from slotqueue The slot queue should also have nothing to do with availabilities * when all availabilities are empty, pause the queue An empty availability is defined as size < DefaultBlockSize as this means even the smallest possible request could not be served. However, this is up for discussion. * remove availability from onAvailabilitiesEmptied callback * refactor onAvailabilityAdded and onAvailabilitiesEmptied onAvailabilityAdded and onAvailabilitiesEmptied are now only called from reservations.update (and eventually reservations.delete once implemented). - Add empty routine for Availability and Reservation - Add allEmpty routine for Availability and Reservation, which returns true when all all Availability or Reservation objects in the datastore are empty. * SlotQueue test support updates * Sales module test support updates * Reservations module tests for queue pausing * Sales module tests for queue pausing Includes tests for sales states cancelled, errored, ignored to ensure onCleanUp is called with correct parameters * SlotQueue module tests for queue pausing * fix existing sales test * PR feedback - indent `self.unpause` - update comment for `clearSeenFlags` * reprocessSlot in SaleErrored only when coming from downloading * remove pausing of queue when availabilities are "emptied" Queue pausing when all availiabilies are "emptied" is not necessary, given that the node would not be able to service slots once all its availabilities' freeSize are too small for the slots in the queue, and would then be paused anyway. Add test that asserts the queue is paused once the freeSpace of availabilities drops too low to fill slots in the queue. * Update clearing of seen flags The asyncheapqueue update overload would need to check index bounds and ultimately a different solution was found using the mitems iterator. * fix test request.id was different before updating request.ask.slots, and that id was used to set the state in mockmarket. * Change filled/cleanup future to nil, so no await is needed * add wait to allow items to be added to queue * do not unpause queue when seen items are pushed * re-add seen item back to queue once paused Previously, when a seen item was processed, it was first popped off the queue, then the queue was paused waiting to process that item once the queue was unpaused. Now, when a seen item is processed, it is popped off the queue, the queue is paused, then the item is re-added to the queue and the queue will wait until unpaused before it will continue popping items off the queue. If the item was not re-added to the queue, it would have been processed immediately once unpaused, however there may have been other items with higher priority pushed to the queue in the meantime. The queue would not be unpaused if those added items were already seen. In particular, this may happen when ignored items due to lack of availability are re-added to a paused queue. Those ignored items will likely have a higher priority than the item that was just seen (due to it having been processed first), causing the queue to the be paused. * address PR comments
2024-05-26 00:38:38 +00:00
return failure(reserveErr.toErr(ReleaseFailedError))
let res = await self.updateImpl(obj)
if oldAvailability.freeSize < obj.freeSize: # availability added
# inform subscribers that Availability has been modified (with increased
# size)
if onAvailabilityAdded =? self.onAvailabilityAdded:
# when chronos v4 is implemented, and OnAvailabilityAdded is annotated
# with async:(raises:[]), we can remove this try/catch as we know, with
# certainty, that nothing will be raised
try:
await onAvailabilityAdded(obj)
except CancelledError as e:
raise e
except CatchableError as e:
# we don't have any insight into types of exceptions that
# `onAvailabilityAdded` can raise because it is caller-defined
warn "Unknown error during 'onAvailabilityAdded' callback", error = e.msg
feat[marketplace]: add slot queue pausing (#752) * add seen flag * Add MockSlotQueueItem and better prioritisation tests * Update seen priority, and include in SlotQueueItem.init * Re-add processed slots to queue Re-add processed slots to queue if the sale was ignored or errored * add pausing of queue - when processing slots in queue, pause queue if item was marked seen - if availability size is increased, trigger onAvailabilityAdded callback - in sales, on availability added, clear 'seen' flags, then unpause the queue - when items pushed to the queue, unpause the queue * remove unused NoMatchingAvailabilityError from slotqueue The slot queue should also have nothing to do with availabilities * when all availabilities are empty, pause the queue An empty availability is defined as size < DefaultBlockSize as this means even the smallest possible request could not be served. However, this is up for discussion. * remove availability from onAvailabilitiesEmptied callback * refactor onAvailabilityAdded and onAvailabilitiesEmptied onAvailabilityAdded and onAvailabilitiesEmptied are now only called from reservations.update (and eventually reservations.delete once implemented). - Add empty routine for Availability and Reservation - Add allEmpty routine for Availability and Reservation, which returns true when all all Availability or Reservation objects in the datastore are empty. * SlotQueue test support updates * Sales module test support updates * Reservations module tests for queue pausing * Sales module tests for queue pausing Includes tests for sales states cancelled, errored, ignored to ensure onCleanUp is called with correct parameters * SlotQueue module tests for queue pausing * fix existing sales test * PR feedback - indent `self.unpause` - update comment for `clearSeenFlags` * reprocessSlot in SaleErrored only when coming from downloading * remove pausing of queue when availabilities are "emptied" Queue pausing when all availiabilies are "emptied" is not necessary, given that the node would not be able to service slots once all its availabilities' freeSize are too small for the slots in the queue, and would then be paused anyway. Add test that asserts the queue is paused once the freeSpace of availabilities drops too low to fill slots in the queue. * Update clearing of seen flags The asyncheapqueue update overload would need to check index bounds and ultimately a different solution was found using the mitems iterator. * fix test request.id was different before updating request.ask.slots, and that id was used to set the state in mockmarket. * Change filled/cleanup future to nil, so no await is needed * add wait to allow items to be added to queue * do not unpause queue when seen items are pushed * re-add seen item back to queue once paused Previously, when a seen item was processed, it was first popped off the queue, then the queue was paused waiting to process that item once the queue was unpaused. Now, when a seen item is processed, it is popped off the queue, the queue is paused, then the item is re-added to the queue and the queue will wait until unpaused before it will continue popping items off the queue. If the item was not re-added to the queue, it would have been processed immediately once unpaused, however there may have been other items with higher priority pushed to the queue in the meantime. The queue would not be unpaused if those added items were already seen. In particular, this may happen when ignored items due to lack of availability are re-added to a paused queue. Those ignored items will likely have a higher priority than the item that was just seen (due to it having been processed first), causing the queue to the be paused. * address PR comments
2024-05-26 00:38:38 +00:00
return res
proc update*(
self: Reservations,
obj: Reservation): Future[?!void] {.async.} =
return await self.updateImpl(obj)
proc update*(
self: Reservations,
obj: Availability): Future[?!void] {.async.} =
withLock(self.availabilityLock):
return await self.updateAvailability(obj)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
proc delete(
self: Reservations,
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
key: Key): Future[?!void] {.async.} =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
trace "deleting object", key
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
if not await self.exists(key):
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return success()
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
if err =? (await self.repo.metaDs.ds.delete(key)).errorOption:
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return failure(err.toErr(DeleteFailedError))
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
return success()
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc deleteReservation*(
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
self: Reservations,
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
reservationId: ReservationId,
availabilityId: AvailabilityId): Future[?!void] {.async.} =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
logScope:
reservationId
availabilityId
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
trace "deleting reservation"
without key =? key(reservationId, availabilityId), error:
return failure(error)
withLock(self.availabilityLock):
without reservation =? (await self.get(key, Reservation)), error:
if error of NotExistsError:
return success()
else:
return failure(error)
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
if reservation.size > 0.u256:
trace "returning remaining reservation bytes to availability",
size = reservation.size
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
without availabilityKey =? availabilityId.key, error:
return failure(error)
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
without var availability =? await self.get(availabilityKey, Availability), error:
return failure(error)
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
availability.freeSize += reservation.size
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
if updateErr =? (await self.updateAvailability(availability)).errorOption:
return failure(updateErr)
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
if err =? (await self.repo.metaDs.ds.delete(key)).errorOption:
return failure(err.toErr(DeleteFailedError))
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return success()
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
feat[marketplace]: add slot queue pausing (#752) * add seen flag * Add MockSlotQueueItem and better prioritisation tests * Update seen priority, and include in SlotQueueItem.init * Re-add processed slots to queue Re-add processed slots to queue if the sale was ignored or errored * add pausing of queue - when processing slots in queue, pause queue if item was marked seen - if availability size is increased, trigger onAvailabilityAdded callback - in sales, on availability added, clear 'seen' flags, then unpause the queue - when items pushed to the queue, unpause the queue * remove unused NoMatchingAvailabilityError from slotqueue The slot queue should also have nothing to do with availabilities * when all availabilities are empty, pause the queue An empty availability is defined as size < DefaultBlockSize as this means even the smallest possible request could not be served. However, this is up for discussion. * remove availability from onAvailabilitiesEmptied callback * refactor onAvailabilityAdded and onAvailabilitiesEmptied onAvailabilityAdded and onAvailabilitiesEmptied are now only called from reservations.update (and eventually reservations.delete once implemented). - Add empty routine for Availability and Reservation - Add allEmpty routine for Availability and Reservation, which returns true when all all Availability or Reservation objects in the datastore are empty. * SlotQueue test support updates * Sales module test support updates * Reservations module tests for queue pausing * Sales module tests for queue pausing Includes tests for sales states cancelled, errored, ignored to ensure onCleanUp is called with correct parameters * SlotQueue module tests for queue pausing * fix existing sales test * PR feedback - indent `self.unpause` - update comment for `clearSeenFlags` * reprocessSlot in SaleErrored only when coming from downloading * remove pausing of queue when availabilities are "emptied" Queue pausing when all availiabilies are "emptied" is not necessary, given that the node would not be able to service slots once all its availabilities' freeSize are too small for the slots in the queue, and would then be paused anyway. Add test that asserts the queue is paused once the freeSpace of availabilities drops too low to fill slots in the queue. * Update clearing of seen flags The asyncheapqueue update overload would need to check index bounds and ultimately a different solution was found using the mitems iterator. * fix test request.id was different before updating request.ask.slots, and that id was used to set the state in mockmarket. * Change filled/cleanup future to nil, so no await is needed * add wait to allow items to be added to queue * do not unpause queue when seen items are pushed * re-add seen item back to queue once paused Previously, when a seen item was processed, it was first popped off the queue, then the queue was paused waiting to process that item once the queue was unpaused. Now, when a seen item is processed, it is popped off the queue, the queue is paused, then the item is re-added to the queue and the queue will wait until unpaused before it will continue popping items off the queue. If the item was not re-added to the queue, it would have been processed immediately once unpaused, however there may have been other items with higher priority pushed to the queue in the meantime. The queue would not be unpaused if those added items were already seen. In particular, this may happen when ignored items due to lack of availability are re-added to a paused queue. Those ignored items will likely have a higher priority than the item that was just seen (due to it having been processed first), causing the queue to the be paused. * address PR comments
2024-05-26 00:38:38 +00:00
# TODO: add support for deleting availabilities
# To delete, must not have any active sales.
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc createAvailability*(
self: Reservations,
size: UInt256,
duration: UInt256,
minPrice: UInt256,
maxCollateral: UInt256): Future[?!Availability] {.async.} =
trace "creating availability", size, duration, minPrice, maxCollateral
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
let availability = Availability.init(
size, size, duration, minPrice, maxCollateral
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
)
let bytes = availability.freeSize.truncate(uint)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
if reserveErr =? (await self.repo.reserve(bytes.NBytes)).errorOption:
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return failure(reserveErr.toErr(ReserveFailedError))
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
if updateErr =? (await self.update(availability)).errorOption:
# rollback the reserve
trace "rolling back reserve"
if rollbackErr =? (await self.repo.release(bytes.NBytes)).errorOption:
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
rollbackErr.parent = updateErr
return failure(rollbackErr)
return failure(updateErr)
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return success(availability)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
method createReservation*(
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
self: Reservations,
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
availabilityId: AvailabilityId,
slotSize: UInt256,
requestId: RequestId,
slotIndex: UInt256
): Future[?!Reservation] {.async, base.} =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
withLock(self.availabilityLock):
without availabilityKey =? availabilityId.key, error:
return failure(error)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
without availability =? await self.get(availabilityKey, Availability), error:
return failure(error)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
# Check that the found availability has enough free space after the lock has been acquired, to prevent asynchronous Availiability modifications
if availability.freeSize < slotSize:
let error = newException(
BytesOutOfBoundsError,
"trying to reserve an amount of bytes that is greater than the total size of the Availability")
return failure(error)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
trace "Creating reservation", availabilityId, slotSize, requestId, slotIndex
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
let reservation = Reservation.init(availabilityId, slotSize, requestId, slotIndex)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
if createResErr =? (await self.update(reservation)).errorOption:
return failure(createResErr)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
# reduce availability freeSize by the slot size, which is now accounted for in
# the newly created Reservation
availability.freeSize -= slotSize
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
# update availability with reduced size
trace "Updating availability with reduced size"
if updateErr =? (await self.updateAvailability(availability)).errorOption:
trace "Updating availability failed, rolling back reservation creation"
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
without key =? reservation.key, keyError:
keyError.parent = updateErr
return failure(keyError)
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
# rollback the reservation creation
if rollbackErr =? (await self.delete(key)).errorOption:
rollbackErr.parent = updateErr
return failure(rollbackErr)
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return failure(updateErr)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
trace "Reservation succesfully created"
return success(reservation)
proc returnBytesToAvailability*(
self: Reservations,
availabilityId: AvailabilityId,
reservationId: ReservationId,
bytes: UInt256): Future[?!void] {.async.} =
logScope:
reservationId
availabilityId
withLock(self.availabilityLock):
without key =? key(reservationId, availabilityId), error:
return failure(error)
without var reservation =? (await self.get(key, Reservation)), error:
return failure(error)
# We are ignoring bytes that are still present in the Reservation because
# they will be returned to Availability through `deleteReservation`.
let bytesToBeReturned = bytes - reservation.size
if bytesToBeReturned == 0:
trace "No bytes are returned", requestSizeBytes = bytes, returningBytes = bytesToBeReturned
return success()
trace "Returning bytes", requestSizeBytes = bytes, returningBytes = bytesToBeReturned
# First lets see if we can re-reserve the bytes, if the Repo's quota
# is depleted then we will fail-fast as there is nothing to be done atm.
if reserveErr =? (await self.repo.reserve(bytesToBeReturned.truncate(uint).NBytes)).errorOption:
return failure(reserveErr.toErr(ReserveFailedError))
without availabilityKey =? availabilityId.key, error:
return failure(error)
without var availability =? await self.get(availabilityKey, Availability), error:
return failure(error)
availability.freeSize += bytesToBeReturned
# Update availability with returned size
if updateErr =? (await self.updateAvailability(availability)).errorOption:
trace "Rolling back returning bytes"
if rollbackErr =? (await self.repo.release(bytesToBeReturned.truncate(uint).NBytes)).errorOption:
rollbackErr.parent = updateErr
return failure(rollbackErr)
return failure(updateErr)
return success()
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc release*(
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
self: Reservations,
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
reservationId: ReservationId,
availabilityId: AvailabilityId,
bytes: uint): Future[?!void] {.async.} =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
logScope:
topics = "release"
bytes
reservationId
availabilityId
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
trace "releasing bytes and updating reservation"
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
without key =? key(reservationId, availabilityId), error:
return failure(error)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
without var reservation =? (await self.get(key, Reservation)), error:
return failure(error)
if reservation.size < bytes.u256:
let error = newException(
BytesOutOfBoundsError,
"trying to release an amount of bytes that is greater than the total size of the Reservation")
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return failure(error)
if releaseErr =? (await self.repo.release(bytes.NBytes)).errorOption:
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return failure(releaseErr.toErr(ReleaseFailedError))
reservation.size -= bytes.u256
# persist partially used Reservation with updated size
if err =? (await self.update(reservation)).errorOption:
# rollback release if an update error encountered
trace "rolling back release"
if rollbackErr =? (await self.repo.reserve(bytes.NBytes)).errorOption:
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
rollbackErr.parent = err
return failure(rollbackErr)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
return failure(err)
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return success()
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
iterator items(self: StorableIter): Future[?seq[byte]] =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
while not self.finished:
yield self.next()
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc storables(
self: Reservations,
T: type SomeStorableObject,
queryKey: Key = ReservationsKey
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
): Future[?!StorableIter] {.async.} =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
var iter = StorableIter()
let query = Query.init(queryKey)
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
when T is Availability:
# should indicate key length of 4, but let the .key logic determine it
without defaultKey =? AvailabilityId.default.key, error:
return failure(error)
elif T is Reservation:
# should indicate key length of 5, but let the .key logic determine it
without defaultKey =? key(ReservationId.default, AvailabilityId.default), error:
return failure(error)
else:
raiseAssert "unknown type"
without results =? await self.repo.metaDs.ds.query(query), error:
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return failure(error)
# /sales/reservations
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc next(): Future[?seq[byte]] {.async.} =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
await idleAsync()
iter.finished = results.finished
if not results.finished and
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
res =? (await results.next()) and
res.data.len > 0 and
key =? res.key and
key.namespaces.len == defaultKey.namespaces.len:
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return some res.data
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return none seq[byte]
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
proc dispose(): Future[?!void] {.async.} =
return await results.dispose()
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
iter.next = next
iter.dispose = dispose
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
return success iter
proc allImpl(
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
self: Reservations,
T: type SomeStorableObject,
queryKey: Key = ReservationsKey
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
): Future[?!seq[T]] {.async.} =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
var ret: seq[T] = @[]
without storables =? (await self.storables(T, queryKey)), error:
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
return failure(error)
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
for storable in storables.items:
without bytes =? (await storable):
continue
without obj =? T.fromJson(bytes), error:
error "json deserialization error",
json = string.fromBytes(bytes),
error = error.msg
continue
ret.add obj
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
return success(ret)
proc all*(
self: Reservations,
T: type SomeStorableObject
): Future[?!seq[T]] {.async.} =
return await self.allImpl(T)
proc all*(
self: Reservations,
T: type SomeStorableObject,
availabilityId: AvailabilityId
): Future[?!seq[T]] {.async.} =
without key =? (ReservationsKey / $availabilityId):
return failure("no key")
return await self.allImpl(T, key)
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
proc findAvailability*(
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
self: Reservations,
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
size, duration, minPrice, collateral: UInt256
): Future[?Availability] {.async.} =
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
without storables =? (await self.storables(Availability)), e:
error "failed to get all storables", error = e.msg
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
return none Availability
[marketplace] Availability improvements (#535) ## Problem When Availabilities are created, the amount of bytes in the Availability are reserved in the repo, so those bytes on disk cannot be written to otherwise. When a request for storage is received by a node, if a previously created Availability is matched, an attempt will be made to fill a slot in the request (more accurately, the request's slots are added to the SlotQueue, and eventually those slots will be processed). During download, bytes that were reserved for the Availability were released (as they were written to disk). To prevent more bytes from being released than were reserved in the Availability, the Availability was marked as used during the download, so that no other requests would match the Availability, and therefore no new downloads (and byte releases) would begin. The unfortunate downside to this, is that the number of Availabilities a node has determines the download concurrency capacity. If, for example, a node creates a single Availability that covers all available disk space the operator is willing to use, that single Availability would mean that only one download could occur at a time, meaning the node could potentially miss out on storage opportunities. ## Solution To alleviate the concurrency issue, each time a slot is processed, a Reservation is created, which takes size (aka reserved bytes) away from the Availability and stores them in the Reservation object. This can be done as many times as needed as long as there are enough bytes remaining in the Availability. Therefore, concurrent downloads are no longer limited by the number of Availabilities. Instead, they would more likely be limited to the SlotQueue's `maxWorkers`. From a database design perspective, an Availability has zero or more Reservations. Reservations are persisted in the RepoStore's metadata, along with Availabilities. The metadata store key path for Reservations is ` meta / sales / reservations / <availabilityId> / <reservationId>`, while Availabilities are stored one level up, eg `meta / sales / reservations / <availabilityId> `, allowing all Reservations for an Availability to be queried (this is not currently needed, but may be useful when work to restore Availability size is implemented, more on this later). ### Lifecycle When a reservation is created, its size is deducted from the Availability, and when a reservation is deleted, any remaining size (bytes not written to disk) is returned to the Availability. If the request finishes, is cancelled (expired), or an error occurs, the Reservation is deleted (and any undownloaded bytes returned to the Availability). In addition, when the Sales module starts, any Reservations that are not actively being used in a filled slot, are deleted. Having a Reservation persisted until after a storage request is completed, will allow for the originally set Availability size to be reclaimed once a request contract has been completed. This is a feature that is yet to be implemented, however the work in this PR is a step in the direction towards enabling this. ### Unknowns Reservation size is determined by the `StorageAsk.slotSize`. If during download, more bytes than `slotSize` are attempted to be downloaded than this, then the Reservation update will fail, and the state machine will move to a `SaleErrored` state, deleting the Reservation. This will likely prevent the slot from being filled. ### Notes Based on #514
2023-09-29 04:33:08 +00:00
for item in storables.items:
if bytes =? (await item) and
availability =? Availability.fromJson(bytes):
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
if size <= availability.freeSize and
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
duration <= availability.duration and
collateral <= availability.maxCollateral and
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
minPrice >= availability.minPrice:
trace "availability matched",
id = availability.id,
size, availFreeSize = availability.freeSize,
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
duration, availDuration = availability.duration,
minPrice, availMinPrice = availability.minPrice,
collateral, availMaxCollateral = availability.maxCollateral
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
# TODO: As soon as we're on ARC-ORC, we can use destructors
# to automatically dispose our iterators when they fall out of scope.
# For now:
if err =? (await storables.dispose()).errorOption:
error "failed to dispose storables iter", error = err.msg
return none Availability
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
return some availability
trace "availability did not match",
id = availability.id,
size, availFreeSize = availability.freeSize,
[marketplace] Add Reservations Module (#340) * [marketplace] reservations module - add de/serialization for Availability - add markUsed/markUnused in persisted availability - add query for unused - add reserve/release - reservation module tests - split ContractInteractions into client contracts and host contracts - remove reservations start/stop as the repo start/stop is being managed by the node - remove dedicated reservations metadata store and use the metadata store from the repo instead - Split ContractInteractions into: - ClientInteractions (with purchasing) - HostInteractions (with sales and proving) - compilation fix for nim 1.2 [repostore] fix started flag, add tests [marketplace] persist slot index For loading the sales state from chain, the slot index was not previously persisted in the contract. Will retrieve the slot index from the contract when the sales state is loaded. * Revert repostore changes In favour of separate PR https://github.com/status-im/nim-codex/pull/374. * remove warnings * clean up * tests: stop repostore during teardown * change constructor type identifier Change Contracts constructor to accept Contracts type instead of ContractInteractions. * change constructor return type to Result instead of Option * fix and split interactions tests * clean up, fix tests * find availability by slot id * remove duplication in host/client interactions * add test for finding availability by slotId * log instead of raiseAssert when failed to mark availability as unused * move to SaleErrored state instead of raiseAssert * remove unneeded reverse It appears that order is not preserved in the repostore, so reversing does not have the intended effect here. * update open api spec for potential rest endpoint errors * move functions about available bytes to repostore * WIP: reserve and release availabilities as needed WIP: not tested yet Availabilities are marked as used when matched (just before downloading starts) so that future matching logic does not match an availability currently in use. As the download progresses, batches of blocks are written to disk, and the equivalent bytes are released from the reservation module. The size of the availability is reduced as well. During a reserve or release operation, availability updates occur after the repo is manipulated. If the availability update operation fails, the reserve or release is rolled back to maintain correct accounting of bytes. Finally, once download completes, or if an error occurs, the availability is marked as unused so future matching can occur. * delete availability when all bytes released * fix tests + cleanup * remove availability from SalesContext callbacks Availability is no longer used past the SaleDownloading state in the state machine. Cleanup of Availability (marking unused) is handled directly in the SaleDownloading state, and no longer in SaleErrored or SaleFinished. Likewise, Availabilities shouldn’t need to be handled on node restart. Additionally, Availability was being passed in SalesContext callbacks, and now that Availability is only used temporarily in the SaleDownloading state, Availability is contextually irrelevant to the callbacks, except in OnStore possibly, though it was not being consumed. * test clean up * - remove availability from callbacks and constructors from previous commit that needed to be removed (oopsie) - fix integration test that checks availabilities - there was a bug fixed that crashed the node due to a missing `return success` in onStore - the test was fixed by ensuring that availabilities are remaining on the node, and the size has been reduced - change Availability back to non-ref object and constructor back to init - add trace logging of all state transitions in state machine - add generally useful trace logging * fixes after rebase 1. Fix onProve callbacks 2. Use Slot type instead of tuple for retrieving active slot. 3. Bump codex-contracts-eth that exposes getActivceSlot call. * swap contracts branch to not support slot collateral Slot collateral changes in the contracts require further changes in the client code, so we’ll skip those changes for now and add in a separate commit. * modify Interactions and Deployment constructors - `HostInteractions` and `ClientInteractions` constructors were simplified to take a contract address and no overloads - `Interactions` prepared simplified so there are no overloads - `Deployment` constructor updated so that it takes an optional string parameter, instead `Option[string]` * Move `batchProc` declaration `batchProc` needs to be consumed by both `node` and `salescontext`, and they can’t reference each other as it creates a circular dependency. * [reservations] rename `available` to `hasAvailable` * [reservations] default error message to inner error msg * add SaleIngored state When a storage request is handled but the request does match availabilities, the sales agent machine is sent to the SaleIgnored state. In addition, the agent is constructed in a way that if the request is ignored, the sales agent is removed from the list of active agents being tracked in the sales module.
2023-04-04 07:05:16 +00:00
duration, availDuration = availability.duration,
minPrice, availMinPrice = availability.minPrice,
collateral, availMaxCollateral = availability.maxCollateral