Open
Conversation
The invariant was broken somehow, but after clean_up returns, the internal state should be consistent again, giving reasonable hope that this will self heal. In reality these should be unreachable though, if the remaining code is respecting the invariant perfectly. Even if the invariant is breached before this, it should not make executing further unsafe.
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 PR fixes a very rare race condition where a peer leaving very close to the expiry of an existing rate limit may leave behind a pending deletion entry while the to be deleted entry is already removed. The main fix is just to make sure that a single timestamp is used for the comparisons, but I also did some mild refactoring to make the code easier to read and understand.
Furthermore I added a regression test which fails before the changes, but does now pass and removed the
unreachable!macros in favor of logging errors as this kind of error does not warrant a full shutdown of the entire client, specifically because the error should be self correcting in these instances where the panic occurred.Resolves #3701