nimbus-eth2/beacon_chain/gossip_processing
Jacek Sieka bd513ca2ba
separate CL/EL timing in add block log (#5380)
the metric remains the same in order not to change overall meaning, but
in the logs we can do better - often, the EL is at the core of a
slowdown.
2023-09-04 20:23:29 +03:00
..
README.md update some consensus-spec URLs to v1.4.0-beta.1 (#5357) 2023-08-25 15:58:44 +00:00
batch_validation.nim fix batch validation when `--num-threads=1` (#5303) 2023-08-17 10:58:13 +02:00
block_processor.nim separate CL/EL timing in add block log (#5380) 2023-09-04 20:23:29 +03:00
eth2_processor.nim ensure blob quarantine insertion always succeeds (#5369) 2023-08-29 16:46:25 +00:00
gossip_validation.nim update some consensus-spec URLs to v1.4.0-beta.1 (#5379) 2023-09-01 09:31:52 +00:00
light_client_processor.nim remove `{.raises: [Defect].}` Nim 1.2 compatibility (#5352) 2023-08-25 11:29:07 +02:00
optimistic_processor.nim remove `{.raises: [Defect].}` Nim 1.2 compatibility (#5352) 2023-08-25 11:29:07 +02:00

README.md

Gossip Processing

This folder holds a collection of modules to:

  • validate raw gossip data before
    • rebroadcasting it (potentially aggregated)
    • sending it to one of the consensus object pools

Validation

Gossip validation is different from consensus verification in particular for blocks.

There are multiple consumers of validated consensus objects:

  • a ValidationResult.Accept output triggers rebroadcasting in libp2p
    • We jump into method validate(PubSub, Message) in libp2p/protocols/pubsub/pubsub.nim
    • which was called by rpcHandler(GossipSub, PubSubPeer, RPCMsg)
  • a blockValidator message enqueues the validated object to the processing queue in block_processor
    • blockQueue: AsyncQueue[BlockEntry] (shared with request_manager and sync_manager)
    • This queue is then regularly processed to be made available to the consensus object pools.
  • a xyzValidator message adds the validated object to a pool in eth2_processor
    • Attestations (unaggregated and aggregated) get collected into batches.
    • Once a threshold is exceeded or after a timeout, they get validated together using BatchCrypto.

Security concerns

As the first line of defense in Nimbus, modules must be able to handle bursts of data that may come:

  • from malicious nodes trying to DOS us
  • from long periods of non-finality, creating lots of forks, attestations