cadonna commented on PR #12875: URL: https://github.com/apache/kafka/pull/12875#issuecomment-2996879605
Both should actually be harmless. The first is just an info because the lock on the state directory of the task could still be taken by a different stream thread on the same Streams client. That might happen if one stream thread wants to initialize the task before another stream thread hasn't closed the same task. The second is written to the log if the lock could not been acquired before the task is moved away from the stream thread. The might happen if two rebalances happen quite fast one after the other. Besides the log messages, do you see other weird behavior like lag that does not decrease for those tasks? Actually, 3.9.1 already contains the backlog mechanism to acquire the log (https://github.com/apache/kafka/pull/17116 and https://github.com/apache/kafka/pull/17209), so I am wondering why you say you see plenty of the first before you see the second. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: jira-unsubscr...@kafka.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
