Stop static group members after fencing - #674
Open
whatyouhide wants to merge 1 commit into
Open
Conversation
Contributor
|
the suggestion was made here: i thought in such situation, the second client will first get fenced, i do not understand how it can cause the full group rebalance. |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
@zmstone I introduced a pretty nasty bug in my static membership work recently: when an instance gets
fenched_instance_id, brod retries to connect. That creates a vicious cycle where the new instance that just joined with the same ID gets itself fenced. The two instances play cat and mouse and eventually cause a full group rebalance, the thing I meant to avoid in the first place with static membership.A note: the reason this happened here even though we've been running this in prod for a while is that the changes we run in our brod fork are not identical. We had originally done some of these changes in broadway_kafka instead, and I backported them here (introducing this bug).
This PR fixes the bug by not retrying
fenced_instance_iderrors. That's what the Java client does too.