Eric 1d161d383e
Slot queue (#455)
## Slot queue
Adds a slot queue, as per the [slot queue design](https://github.com/codex-storage/codex-research/blob/master/design/sales.md#slot-queue).

Any time storage is requested, all slots from that request are immediately added to the queue. Finished, Canclled, Failed requests remove all slots with that request id from the queue. SlotFreed events add a new slot to the queue and SlotFilled events remove the slot from the queue. This allows popping of a slot each time one is processed, making things much simpler.

When an entire request of slots is added to the queue, the slot indices are shuffled randomly to hopefully prevent nodes that pick up the same storage requested event from clashing on the first processed slot index. This allowed removal of assigning a random slot index in the SalePreparing state and it also ensured that all SalesAgents will have a slot index assigned to them at the start thus the removal of the optional slotIndex.

Remove slotId from SlotFreed event as it was not being used. RequestId and slotIndex were added to the SlotFreed event earlier and those are now being used

The slot queue invariant that prioritises queue items added to the queue relies on a scoring mechanism to sort them based on the [sort order in the design document](https://github.com/codex-storage/codex-research/blob/master/design/sales.md#sort-order).

When a storage request is handled by the sales module, a slot index was randomly assigned and then the slot was filled. Now, a random slot index is only assigned when adding an entire request to the slot queue. Additionally, the slot is checked that its state is `SlotState.Free` before continuing with the download process.

SlotQueue should always ensure the underlying AsyncHeapQueue has one less than the maximum items, ensuring the SlotQueue can always have space to add an additional item regardless if it’s full or not.

Constructing `SlotQueue.workers` in `SlotQueue.new` calls `newAsyncQueue` which causes side effects, so the construction call had to be moved to `SlotQueue.start`.

Prevent loading request from contract (network request) if there is an existing item in queue for that request.

Check availability before adding request to queue.

Add ability to query market contract for past events. When new availabilities are added, the `onReservationAdded` callback is triggered in which past `StorageRequested` events are queried, and those slots are added to the queue (filtered by availability on `push` and filtered by state in `SalePreparing`).

#### Request Workers
Limit the concurrent requests being processed in the queue by using a limited pool of workers (default = 3). Workers are in a data structure of type `AsyncQueue[SlotQueueWorker]`. This allows us to await a `popFirst` for available workers inside of the main SlotQueue event loop

Add an `onCleanUp` that stops the agents and removes them from the sales module agent list. `onCleanUp` is called from sales end states (eg ignored, cancelled, finished, failed, errored).

Add a `doneProcessing` future to `SlotQueueWorker` to be completed in the `OnProcessSlot` callback. Each `doneProcessing` future created is cancelled and awaited in `SlotQueue.stop` (thanks to `TrackableFuturees`), which forced `stop` to become async.
  - Cancel dispatched workers and the `onProcessSlot` callbacks, prevents zombie callbacks

#### Add TrackableFutures
Allow tracking of futures in a module so they can be cancelled at a later time. Useful for asyncSpawned futures, but works for any future.

### Sales module
The sales module needed to subscribe to request events to ensure that the request queue was managed correctly on each event. In the process of doing this, the sales agents were updated to avoid subscribing to events in each agent, and instead dispatch received events from the sales module to all created sales agents. This would prevent memory leaks on having too many eventemitters subscribed to.
  - prevent removal of agents from sales module while stopping, otherwise the agents seq len is modified while iterating

An additional sales agent state was added, `SalePreparing`, that handles all state machine setup, such as retrieving the request and subscribing to events that were previously in the `SaleDownloading` state.

Once agents have parked in an end state (eg ignored, cancelled, finished, failed, errored), they were not getting cleaned up and the sales module was keeping a handle on their reference. An `onCleanUp` callback was created to be called after the state machine enters an end state, which could prevent a memory leak if the number of requests coming in is high.

Move the SalesAgent callback raises pragmas from the Sales module to the proc definition in SalesAgent. This avoids having to catch `Exception`.
  - remove unneeded error handling as pragmas were moved

Move sales.subscriptions from an object containing named subscriptions to a `seq[Subscription]` directly on the sales object.

Sales tests: shut down repo after sales stop, to fix SIGABRT in CI

### Add async Promise API
  - modelled after JavaScript Promise API
  - alternative to `asyncSpawn` that allows handling of async calls in a synchronous context (including access to the synchronous closure) with less additional procs to be declared
  - Write less code, catch errors that would otherwise defect in asyncspawn, and execute a callback after completion
  - Add cancellation callbacks to utils/then, ensuring cancellations are handled properly

## Dependencies
- bump codex-contracts-eth to support slot queue (https://github.com/codex-storage/codex-contracts-eth/pull/61)
- bump nim-ethers to 0.5.0
- Bump nim-json-rpc submodule to 0bf2bcb

---------

Co-authored-by: Jaremy Creechley <creechley@gmail.com>
2023-07-25 12:50:30 +10:00

408 lines
12 KiB
Nim

import std/sequtils
import std/sugar
import std/tables
import pkg/questionable
import pkg/stint
import pkg/chronicles
import pkg/datastore
import ./market
import ./clock
import ./proving
import ./stores
import ./contracts/requests
import ./contracts/marketplace
import ./sales/salescontext
import ./sales/salesagent
import ./sales/statemachine
import ./sales/slotqueue
import ./sales/trackedfutures
import ./sales/states/preparing
import ./sales/states/unknown
import ./utils/then
## Sales holds a list of available storage that it may sell.
##
## When storage is requested on the market that matches availability, the Sales
## object will instruct the Codex node to persist the requested data. Once the
## data has been persisted, it uploads a proof of storage to the market in an
## attempt to win a storage contract.
##
## Node Sales Market
## | | |
## | -- add availability --> | |
## | | <-- storage request --- |
## | <----- store data ------ | |
## | -----------------------> | |
## | | |
## | <----- prove data ---- | |
## | -----------------------> | |
## | | ---- storage proof ---> |
export stint
export reservations
logScope:
topics = "sales"
type
Sales* = ref object
context*: SalesContext
agents*: seq[SalesAgent]
running: bool
subscriptions: seq[market.Subscription]
trackedFutures: TrackedFutures
proc `onStore=`*(sales: Sales, onStore: OnStore) =
sales.context.onStore = some onStore
proc `onClear=`*(sales: Sales, onClear: OnClear) =
sales.context.onClear = some onClear
proc `onSale=`*(sales: Sales, callback: OnSale) =
sales.context.onSale = some callback
proc onStore*(sales: Sales): ?OnStore = sales.context.onStore
proc onClear*(sales: Sales): ?OnClear = sales.context.onClear
proc onSale*(sales: Sales): ?OnSale = sales.context.onSale
func new*(_: type Sales,
market: Market,
clock: Clock,
proving: Proving,
repo: RepoStore): Sales =
let reservations = Reservations.new(repo)
Sales(
context: SalesContext(
market: market,
clock: clock,
proving: proving,
reservations: reservations,
slotQueue: SlotQueue.new(reservations)
),
trackedFutures: TrackedFutures.new(),
subscriptions: @[]
)
proc remove(sales: Sales, agent: SalesAgent) {.async.} =
await agent.stop()
if sales.running:
sales.agents.keepItIf(it != agent)
proc cleanUp(sales: Sales,
agent: SalesAgent,
processing: Future[void]) {.async.} =
await sales.remove(agent)
# signal back to the slot queue to cycle a worker
if not processing.isNil and not processing.finished():
processing.complete()
proc processSlot(sales: Sales, item: SlotQueueItem, done: Future[void]) =
debug "processing slot from queue", requestId = $item.requestId,
slot = item.slotIndex
let agent = newSalesAgent(
sales.context,
item.requestId,
item.slotIndex.u256,
none StorageRequest
)
agent.context.onCleanUp = proc {.async.} =
await sales.cleanUp(agent, done)
agent.start(SalePreparing())
sales.agents.add agent
proc mySlots*(sales: Sales): Future[seq[Slot]] {.async.} =
let market = sales.context.market
let slotIds = await market.mySlots()
var slots: seq[Slot] = @[]
for slotId in slotIds:
if slot =? (await market.getActiveSlot(slotId)):
slots.add slot
return slots
proc load*(sales: Sales) {.async.} =
let slots = await sales.mySlots()
for slot in slots:
let agent = newSalesAgent(
sales.context,
slot.request.id,
slot.slotIndex,
some slot.request)
agent.context.onCleanUp = proc {.async.} = await sales.remove(agent)
agent.start(SaleUnknown())
sales.agents.add agent
proc onReservationAdded(sales: Sales, availability: Availability) {.async.} =
## Query last 256 blocks for new requests, adding them to the queue. `push`
## checks for availability before adding to the queue. If processed, the
## sales agent will check if the slot is free.
let context = sales.context
let market = context.market
let queue = context.slotQueue
logScope:
topics = "sales onReservationAdded callback"
trace "reservation added, querying past storage requests to add to queue"
try:
let events = await market.queryPastStorageRequests(256)
let requests = events.map(event =>
SlotQueueItem.init(event.requestId, event.ask, event.expiry)
)
trace "found past storage requested events to add to queue",
events = events.len
for slots in requests:
for slot in slots:
if err =? (await queue.push(slot)).errorOption:
# continue on error
if err of QueueNotRunningError:
warn "cannot push items to queue, queue is not running"
elif err of NoMatchingAvailabilityError:
info "slot in queue had no matching availabilities, ignoring"
elif err of SlotsOutOfRangeError:
warn "Too many slots, cannot add to queue"
elif err of SlotQueueItemExistsError:
trace "item already exists, ignoring"
discard
else: raise err
except CatchableError as e:
warn "Error adding request to SlotQueue", error = e.msg
discard
proc onStorageRequested(sales: Sales,
requestId: RequestId,
ask: StorageAsk,
expiry: UInt256) =
logScope:
topics = "sales onStorageRequested"
requestId
slots = ask.slots
expiry
let slotQueue = sales.context.slotQueue
trace "storage requested, adding slots to queue"
without items =? SlotQueueItem.init(requestId, ask, expiry).catch, err:
if err of SlotsOutOfRangeError:
warn "Too many slots, cannot add to queue"
else:
warn "Failed to create slot queue items from request", error = err.msg
for item in items:
# continue on failure
slotQueue.push(item)
.track(sales)
.catch(proc(err: ref CatchableError) =
if err of NoMatchingAvailabilityError:
info "slot in queue had no matching availabilities, ignoring"
elif err of SlotQueueItemExistsError:
error "Failed to push item to queue becaue it already exists"
elif err of QueueNotRunningError:
warn "Failed to push item to queue becaue queue is not running"
else:
warn "Error adding request to SlotQueue", error = err.msg
)
proc onSlotFreed(sales: Sales,
requestId: RequestId,
slotIndex: UInt256) =
logScope:
topics = "sales onSlotFreed"
requestId
slotIndex
trace "slot freed, adding to queue"
proc addSlotToQueue() {.async.} =
let context = sales.context
let market = context.market
let queue = context.slotQueue
# first attempt to populate request using existing slot metadata in queue
without var found =? queue.populateItem(requestId,
slotIndex.truncate(uint16)):
trace "no existing request metadata, getting request info from contract"
# if there's no existing slot for that request, retrieve the request
# from the contract.
without request =? await market.getRequest(requestId):
error "unknown request in contract"
return
found = SlotQueueItem.init(request, slotIndex.truncate(uint16))
if err =? (await queue.push(found)).errorOption:
raise err
addSlotToQueue()
.track(sales)
.catch(proc(err: ref CatchableError) =
if err of NoMatchingAvailabilityError:
info "slot in queue had no matching availabilities, ignoring"
elif err of SlotQueueItemExistsError:
error "Failed to push item to queue becaue it already exists"
elif err of QueueNotRunningError:
warn "Failed to push item to queue becaue queue is not running"
else:
warn "Error adding request to SlotQueue", error = err.msg
)
proc subscribeRequested(sales: Sales) {.async.} =
let context = sales.context
let market = context.market
proc onStorageRequested(requestId: RequestId,
ask: StorageAsk,
expiry: UInt256) =
sales.onStorageRequested(requestId, ask, expiry)
try:
let sub = await market.subscribeRequests(onStorageRequested)
sales.subscriptions.add(sub)
except CatchableError as e:
error "Unable to subscribe to storage request events", msg = e.msg
proc subscribeCancellation(sales: Sales) {.async.} =
let context = sales.context
let market = context.market
let queue = context.slotQueue
proc onCancelled(requestId: RequestId) =
trace "request cancelled, removing all request slots from queue"
queue.delete(requestId)
try:
let sub = await market.subscribeRequestCancelled(onCancelled)
sales.subscriptions.add(sub)
except CatchableError as e:
error "Unable to subscribe to cancellation events", msg = e.msg
proc subscribeFulfilled*(sales: Sales) {.async.} =
let context = sales.context
let market = context.market
let queue = context.slotQueue
proc onFulfilled(requestId: RequestId) =
trace "request fulfilled, removing all request slots from queue"
queue.delete(requestId)
for agent in sales.agents:
agent.onFulfilled(requestId)
try:
let sub = await market.subscribeFulfillment(onFulfilled)
sales.subscriptions.add(sub)
except CatchableError as e:
error "Unable to subscribe to storage fulfilled events", msg = e.msg
proc subscribeFailure(sales: Sales) {.async.} =
let context = sales.context
let market = context.market
let queue = context.slotQueue
proc onFailed(requestId: RequestId) =
trace "request failed, removing all request slots from queue"
queue.delete(requestId)
for agent in sales.agents:
agent.onFailed(requestId)
try:
let sub = await market.subscribeRequestFailed(onFailed)
sales.subscriptions.add(sub)
except CatchableError as e:
error "Unable to subscribe to storage failure events", msg = e.msg
proc subscribeSlotFilled(sales: Sales) {.async.} =
let context = sales.context
let market = context.market
let queue = context.slotQueue
proc onSlotFilled(requestId: RequestId, slotIndex: UInt256) =
trace "slot filled, removing from slot queue", requestId, slotIndex
queue.delete(requestId, slotIndex.truncate(uint16))
for agent in sales.agents:
agent.onSlotFilled(requestId, slotIndex)
try:
let sub = await market.subscribeSlotFilled(onSlotFilled)
sales.subscriptions.add(sub)
except CatchableError as e:
error "Unable to subscribe to slot filled events", msg = e.msg
proc subscribeSlotFreed(sales: Sales) {.async.} =
let context = sales.context
let market = context.market
proc onSlotFreed(requestId: RequestId, slotIndex: UInt256) =
sales.onSlotFreed(requestId, slotIndex)
try:
let sub = await market.subscribeSlotFreed(onSlotFreed)
sales.subscriptions.add(sub)
except CatchableError as e:
error "Unable to subscribe to slot freed events", msg = e.msg
proc startSlotQueue(sales: Sales) {.async.} =
let slotQueue = sales.context.slotQueue
let reservations = sales.context.reservations
slotQueue.onProcessSlot =
proc(item: SlotQueueItem, done: Future[void]) {.async.} =
sales.processSlot(item, done)
asyncSpawn slotQueue.start()
reservations.onReservationAdded =
proc(availability: Availability) {.async.} =
await sales.onReservationAdded(availability)
proc subscribe(sales: Sales) {.async.} =
await sales.subscribeRequested()
await sales.subscribeFulfilled()
await sales.subscribeFailure()
await sales.subscribeSlotFilled()
await sales.subscribeSlotFreed()
await sales.subscribeCancellation()
proc unsubscribe(sales: Sales) {.async.} =
for sub in sales.subscriptions:
try:
await sub.unsubscribe()
except CatchableError as e:
error "Unable to unsubscribe from subscription", error = e.msg
proc start*(sales: Sales) {.async.} =
await sales.startSlotQueue()
await sales.subscribe()
proc stop*(sales: Sales) {.async.} =
trace "stopping sales"
sales.running = false
await sales.context.slotQueue.stop()
await sales.unsubscribe()
await sales.trackedFutures.cancelTracked()
for agent in sales.agents:
await agent.stop()
sales.agents = @[]