You are viewing a plain text version of this content. The canonical link for it is here.

Posted to jira@kafka.apache.org by GitBox <gi...@apache.org> on 2020/05/13 21:35:28 UTC

[GitHub] [kafka] junrao commented on pull request #8479: KAFKA-9769: Finish operations for leaderEpoch-updated partitions up to point ZK Exception

junrao commented on pull request #8479:
URL: https://github.com/apache/kafka/pull/8479#issuecomment-628258485


   @andrewchoi5 : Thanks for finding this issue. We fixed https://issues.apache.org/jira/browse/KAFKA-9932 recently. So, the `fetchLogConfig` from ZK method is called rarely during makeFollowers() now. If we do hit the ZK exception, there are a few options: (1) As Jason mentioned, we could keep retrying from ZK until successful. This probably needs to be done in a separate thread to avoid blocking the request handler thread. So, it can be a bit involved. (2) Don't update the leader epoch (we can choose to update the leader epoch after the local log is obtained successfully) and log an error. Since this issue should be rare now, maybe we can do (2) for now?


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org