Use RWMutex in demultiplexer to prevent check delays #26148
Merged
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Outdated suggestions cannot be applied.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.
What does this PR do?
Followup to this PR: #25858
Very long running flushes can cause the lock in
demultiplexer_agent
to be held while checks are starting. When checks start they callGetSender
to acquire their sender instance which shares the same lock as flush. As a result, a very long running flush will block all checks from starting until it completes.In order to avoid check execution delays, we can use an RWMutex here since the protected state is only mutated during
Stop
. SoRLock
will never block during normal agent operations allowing checks to start asynchronously of the flush operation.Motivation
Additional Notes
Possible Drawbacks / Trade-offs
Describe how to test/QA your changes
This was deployed in a large staging cluster so QA is done.