feat(code/blocksync): Randomize choice of peer to sync from #492
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## blocksync-mvp #492 +/- ##
=================================================
+ Coverage 78.70% 78.81% +0.11%
=================================================
Files 124 124
Lines 8715 8735 +20
=================================================
+ Hits 6859 6884 +25
+ Misses 1856 1851 -5
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. |
romac
added a commit
that referenced
this pull request
Oct 30, 2024
* feat(code): Add dummy `BlockSync` actor * Add `/block_sync` pubsub channel * Rename `broadcast` to `publish` * Use `eyre` instead of boxed `Error` * Have peers publish their status every time they move to a new height/round * Fix bug where `StartRound` effect was performed twice * Move status handling into BlockSync actor * Less noise * Move status update logic onto BlockSync actor * Decouple GossipConsensus actor from Consensus and BlockSync actors * Push latest hight from consensus to blocksync * Revert "Decouple GossipConsensus actor from Consensus and BlockSync actors" This reverts commit 0a59ada. * feat(code): BlockSync RPC (#457) * Define RPC behaviour for BlockSync * Hook BlockSync behaviour into networking layer * Revert to encoding-agnostic networking layer * Hook BlockSync actor to BlockSync behaviour * Fix unwraps * Remove round from status * code(feat): Add block store (#458) * Add block store for blocksync testing * Sync helper get stored block at height from host * Add block store pruning * Update Cargo.lock * Add blocksync protocol initial implementation, wip * Add config flag for maximum number of blocks in store * Ask for the block only from peer at lower height. * Fix mistake in host, should return ReceivedProposedValue and not ProposedValue. * When processing the synced block use the on_ handlers instead of apply_driver_input. * When storing the decision look for full proposal at proposal round and NOT consensus round * Change max_retain_blocks in config.toml to match the default. * Set voting power for all vals to 1 for easier debugging. * Use on_vote instead of apply_driver_input so our own commit is stored and used on decided operations. * Fix spelling. * Remove lower number of peers (was used for testing). * Store the commits in a set to avoid duplicates. * Return if proposal from non-proposer. * Add TODO for potential min block height in Status * Change max_retain_blocks default to 100. * Small cleanup * Work around libp2p bug where a node cannot publish to a topic if restarted too quickly See libp2p/rust-libp2p#5097 * Cleanup * Ensure validator set is always sorted properly, even after being deserialized from genesis * Update spawn.fish * Move BlockSyncState to blocksync crate * Add batched request for blocks * Fix bug where only a single block would be sent back * Revert block batching * Cleanup * Remove outdated comment * Post-merge fixes * Post-merge fix * Cleanup * chore(code): Extract logic from `BlockSync` actor and into `blocksync` crate (#487) * chore(code): Extract BlockSync logic from actor and into `blocksync` crate * Cleanup * Fix doc * Fix logs * Use custom timeout per test * Add timeouts to BlockSync requests (#490) * feat(code/blocksync): Add timeouts to BlockSync requests * Fix parsing of blocksync config * Randomize choice of peer to sync from (#492) * Add basic integration test (#493) * Re-enable mempool in integration test * Add config option `blocksync.enabled` * Send back actual block bytes instead of just the block hash * Fix example config file * test(code/blocksync): Introduce small DSL for expressing test scenarios (#496) * Introduce small DSL for expressing test scenarios * Add basic integration test for BlockSync * Update fish script * Allow overriding the transport used in tests via a `MALACHITE_TRANSPORT` env variable * Update test * Add doc comments * Use TCP by default in tests Can be overriden locally with `MALACHITE_TRANSPORT=quic` * Run integration tests sequentially * Run integration tests in release mode * Fix coverage job * Try again * Do not panic when tracing subscribers are set twice * Enable basic optimizations in debug mode * Update MSRV to 1.82 * feat(code/blocksync): Only request sync from peers which have the requested block in their store (#495) * feat(code/blocksync): Only request sync from peers which have the requested block in their store For this, we extend the status with the earliest block height available in the store, to ensure we only request a block from peers which have told us they have it. * Remove `blocksync::State::random_peer_at_or_above()` method * Update clippy msrv * Add metrics * Ensure Consensus and BlockSync actors never fail --------- Co-authored-by: Anca Zamfir <ancazamfir@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.