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 2020/07/12 19:10:22 UTC

[GitHub] [kafka] guozhangwang opened a new pull request #9011: KAFKA-10134: Still depends on existence of any fetchable partitions to block on join

guozhangwang opened a new pull request #9011:
URL: https://github.com/apache/kafka/pull/9011


   
   
   ### Committer Checklist (excluded from commit message)
   - [ ] Verify design and implementation 
   - [ ] Verify test coverage and CI build status
   - [ ] Verify documentation (including upgrade notes)
   


----------------------------------------------------------------
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

[GitHub] [kafka] guozhangwang closed pull request #9011: KAFKA-10134: Still depends on existence of any fetchable partitions to block on join

Posted by GitBox <gi...@apache.org>.

guozhangwang closed pull request #9011:
URL: https://github.com/apache/kafka/pull/9011


   


----------------------------------------------------------------
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

[GitHub] [kafka] hachikuji commented on pull request #9011: KAFKA-10134: Still depends on existence of any fetchable partitions to block on join

Posted by GitBox <gi...@apache.org>.

hachikuji commented on pull request #9011:
URL: https://github.com/apache/kafka/pull/9011#issuecomment-659560017


   @guozhangwang There is one detail I think we're missing. If `updateAssignmentMetadataIfNeeded` does not block, then execution will fall through to `pollForFetches`. I would like to understand why `pollForFetches` is not blocking. As far as I can tell, the only thing that would cause that is if `Heartbeat.timeToNextHeartbeat` is returning 0. 


----------------------------------------------------------------
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

[GitHub] [kafka] guozhangwang edited a comment on pull request #9011: KAFKA-10134: Still depends on existence of any fetchable partitions to block on join

Posted by GitBox <gi...@apache.org>.

guozhangwang edited a comment on pull request #9011:
URL: https://github.com/apache/kafka/pull/9011#issuecomment-665182983


   @hachikuji That's what I was wondering as well. From the logs we have: https://issues.apache.org/jira/secure/attachment/13008127/consumer5.log.2020-07-22.log
   
   If you search for `Returning timer remaining` you can find that in most time it is normal, and during that re-joining it is indeed returning 0 but that's expected since we could not connect to the broker during that period of time. So there's no surprising `timeToNextHeartbeat` returning 0 from that run.


----------------------------------------------------------------
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

[GitHub] [kafka] guozhangwang commented on pull request #9011: KAFKA-10134: Still depends on existence of any fetchable partitions to block on join

Posted by GitBox <gi...@apache.org>.

guozhangwang commented on pull request #9011:
URL: https://github.com/apache/kafka/pull/9011#issuecomment-665182983


   That's what I was wondering as well. From the logs we have: https://issues.apache.org/jira/secure/attachment/13008127/consumer5.log.2020-07-22.log
   
   If you search for `Returning timer remaining` you can find that in most time it is normal, and during that re-joining it is indeed returning 0 but that's expected since we could not connect to the broker during that period of time.


----------------------------------------------------------------
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