Fix OurViewChange small race #5356
Merged
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
Signed-off-by: Alexandru Gheorghe <[email protected]>
eskimor
approved these changes
Aug 14, 2024
ordian
approved these changes
Aug 14, 2024
sandreim
approved these changes
Aug 14, 2024
Signed-off-by: Alexandru Gheorghe <[email protected]>
alexggh
added
T0-node
So the message traveled faster via TCP to a different peer and this answered us faster with a new message that got processed before the ourviewchange? And wouldn't the rx side also be first required to process these messages before they are propagated to the subsystem? |
ordian
added a commit
that referenced
this pull request
Aug 14, 2024
* master: (35 commits) Fix OurViewChange small race (#5356) Make ticket non-optional and add ensure_successful method to Consideration trait (#5359) [tests] dedup test code, add more tests, improve naming and docs (#5338) Stop running the wishlist workflow on forks (#5297) Migrate foreign assets v3::Location to v4::Location (#4129) Minor clean up (#5284) [Pools] Ensure members can always exit the pool gracefully (#4998) StorageWeightReclaim: set to node pov size if higher (#5281) [Bot] Add prdoc generation (#5331) Small nits found accidentally along the way (#5341) Create subsystem-benchmarks.yml (#5325) Bump libp2p-identity from 0.2.8 to 0.2.9 (#5232) Bump authoring duration for async backing to 2s. (#5195) Fix spelling issues (#5206) Bump the known_good_semver group across 1 directory with 3 updates (#5315) `polkadot-node-core-pvf-common`: Fix test compilation error (#5310) ci: Paused `cmd-action` commenter (#5287) Remove unnecessary mut (#5318) chain-spec: minor clarification on the genesis config patch (#5324) fix av-distribution Jaeger spans mem leak (#5321) ...
ordian
added a commit
that referenced
this pull request
Aug 14, 2024
* master: (35 commits) Fix OurViewChange small race (#5356) Make ticket non-optional and add ensure_successful method to Consideration trait (#5359) [tests] dedup test code, add more tests, improve naming and docs (#5338) Stop running the wishlist workflow on forks (#5297) Migrate foreign assets v3::Location to v4::Location (#4129) Minor clean up (#5284) [Pools] Ensure members can always exit the pool gracefully (#4998) StorageWeightReclaim: set to node pov size if higher (#5281) [Bot] Add prdoc generation (#5331) Small nits found accidentally along the way (#5341) Create subsystem-benchmarks.yml (#5325) Bump libp2p-identity from 0.2.8 to 0.2.9 (#5232) Bump authoring duration for async backing to 2s. (#5195) Fix spelling issues (#5206) Bump the known_good_semver group across 1 directory with 3 updates (#5315) `polkadot-node-core-pvf-common`: Fix test compilation error (#5310) ci: Paused `cmd-action` commenter (#5287) Remove unnecessary mut (#5318) chain-spec: minor clarification on the genesis config patch (#5324) fix av-distribution Jaeger spans mem leak (#5321) ...
Yes, that was the chronology from timestamps on versi where the CPU is 8 times over-committed.
Yes they would come through rx, I guess your question is where is the context switch between this task and the rx one, right ? |
ordian
added a commit
that referenced
this pull request
Aug 15, 2024
* master: (167 commits) Upgrade accidentally downgraded deps (#5365) [Pools] Fix issues with member migration to `DelegateStake` (#4822) Unify `no_genesis` check (#5360) [CI] Fix prdoc command (#5358) Beefy: add benchmarks for `report_fork_voting()` (#5188) Fix OurViewChange small race (#5356) Make ticket non-optional and add ensure_successful method to Consideration trait (#5359) [tests] dedup test code, add more tests, improve naming and docs (#5338) Stop running the wishlist workflow on forks (#5297) Migrate foreign assets v3::Location to v4::Location (#4129) Minor clean up (#5284) [Pools] Ensure members can always exit the pool gracefully (#4998) StorageWeightReclaim: set to node pov size if higher (#5281) [Bot] Add prdoc generation (#5331) Small nits found accidentally along the way (#5341) Create subsystem-benchmarks.yml (#5325) Bump libp2p-identity from 0.2.8 to 0.2.9 (#5232) Bump authoring duration for async backing to 2s. (#5195) Fix spelling issues (#5206) Bump the known_good_semver group across 1 directory with 3 updates (#5315) ...
Just seen that the orchestra and network event handler are two different tasks. Still crazy that this worked, but explains how it could work. Ty! |
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.
Always queue OurViewChange event before we send view changes to our peers, because otherwise we risk the peers sending us a message that can be processed by our subsystems before OurViewChange.
Normally, this is not really a problem because the latency of the ViewChange we send to our peers is way higher that our subsystem processing OurViewChange, however on testnets like versi where CPU is sometimes overcommitted this race gets triggered occasionally, so let's fix it by sending the messages in the right order.