Merged
Conversation
Member
Author
|
No CHANGELOG entry is required. |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #148 +/- ##
==========================================
+ Coverage 57.66% 57.71% +0.04%
==========================================
Files 33 33
Lines 2277 2282 +5
==========================================
+ Hits 1313 1317 +4
- Misses 875 876 +1
Partials 89 89 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Member
Author
|
OK, something is not clear with this upgrade, consensus service is hanging (even if dBFT master is used). |
Member
Author
|
OK, it looks like we still need to drain the timer's channel, otherwise consensus doesn't work properly. |
Member
|
Just do nspcc-dev/neo-go#3553, this is not needed. |
Member
Author
|
I still think that timer draining related commit should be reverted because there's a possibility that timer is requested to be reset with Lines 157 to 159 in 5281a9a Consider the situation when CV happens: the timer will be reset in ChangeView message handler: Line 179 in 5281a9a Which means that dBFT eventloop is still busy with CV handling and can't read from the timer's channel (given that Lines 60 to 65 in 5281a9a At the same time, as it is said in the timer's documentation, this channel isn't buffered anymore:
So we end up in a deadlock. Take a look at the test failing in nspcc-dev/neo-go@d60f35c, or, as an example, run privnet from the same commit, kill one node and see the result: nodes won't be able to process CV properly: |
This reverts commit f7a23ed. Draining is needed because there's a possibility that timer is requested to be reset with `0` delta: https://github.com/nspcc-dev/dbft/blob/5281a9ae584e866e6dfa2a99313f7db6818c8e9d/dbft.go#L157-L159 Consider the situation when CV happens: the timer will be reset in ChangeView message handler: https://github.com/nspcc-dev/dbft/blob/5281a9ae584e866e6dfa2a99313f7db6818c8e9d/check.go#L179 Which means that dBFT eventloop is still busy with CV handling and can't read from the timer's channel (given that `d` is `0`): https://github.com/nspcc-dev/dbft/blob/5281a9ae584e866e6dfa2a99313f7db6818c8e9d/timer/timer.go#L60-L65 And this channel has a buffer of `1` in case if it's internal Timer-bound channel, which is not enough to finish CV processing. So we end up in a deadlock. Signed-off-by: Anna Shaleva <shaleva.ann@nspcc.ru>
roman-khimov
approved these changes
Mar 21, 2025
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
This reverts commit 8dfda6b and f7a23ed.
This change is needed because external users of the library should be able to stop the dBFT timer:
https://github.com/nspcc-dev/neo-go/blob/3e54c46281237124da7fd89a90501a9d58a3aa60/pkg/consensus/consensus.go#L342