[server] Avoid race condition between pick consumer call and actual consumer subscribe API #1301
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
…onsumer subscribe API
gaojieliu
approved these changes
Nov 13, 2024
There was a problem hiding this comment.
This code change is fine overall, but it is not fully accurate as unsubscription can happen concurrently with pickConsumerForPartition call.
Since pickConsumerForPartition function is tracking these maps syncly, it can guarantee at least the even distribution for the same resource, which is potentially good enough.
If ^ is a correct understanding, can you convert it into javadoc to avoid confusion.
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Outdated suggestions cannot be applied.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.
[server] Avoid race condition between pick consumer call and actual consumer subscribe API
Current implementation does not cover the race condition case between pick consumer API call and actual consumer subscribe call. So the assignment size can be inaccurate during calculation. To address this issue, this PR adds maps to keep track of the basic load and store level load for each consumer, and update it during pickConsumerForPartition() API
How was this PR tested?
Adjust unit test to more coverage.
Does this PR introduce any user-facing changes?