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 2022/09/21 15:14:43 UTC

[GitHub] [kafka] C0urante commented on pull request #12671: KAFKA-14250: MirrorSourceTask exception causes the task to fail

C0urante commented on PR #12671:
URL: https://github.com/apache/kafka/pull/12671#issuecomment-1253855738

   @viktorsomogyi I'm not sure I understand the motivation for this change yet.
   
   The only time the offset reader should only be closed while the `MirrorSourceTask` is still running is when the task is being cancelled because it has been scheduled for shutdown by the Kafka Connect framework, but has not completed that shutdown in time. When tasks are cancelled, we intentionally refrain from emitting statuses for them as there may be new instances of the same task already up and running. So, we shouldn't be seeing inaccurate task statuses as a result of an offset read by a `MirrorSourceTask` being interrupted during startup.
   
   Even when a task is cancelled, we make sure to invoke `Task::stop` on it as soon as we're able to, which should allow the task to clean up its resources.
   
   I'm also unclear on what the ideal behavior is with regards to `FAILED` vs. `STOPPED`; it sounds like there's an expectation that the task gets automatically restarted without user intervention (e.g., hitting the `/connectors/{connector}/tasks/{task}/restart` endpoint). Can you shed some light on what kind of user-facing behavior you expect to see with Kafka Connect / MirrorMaker 2 with this change?


-- 
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-unsubscribe@kafka.apache.org

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