[
https://issues.apache.org/jira/browse/MAPREDUCE-4055?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13235586#comment-13235586
]
Robert Joseph Evans commented on MAPREDUCE-4055:
------------------------------------------------
I believe, and could very well be wrong, that the problem here is that some of
the the threads that do cleanup etc in the history server when they catch an
exception (like when the NameNode is down) they simply log the exception and
exit the thread. Once the thread has exited we will not try any more to clean
up or move data to done.
> Job history files are not getting copied from "intermediate done" directory
> to "done" directory
> ------------------------------------------------------------------------------------------------
>
> Key: MAPREDUCE-4055
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-4055
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: mrv2
> Reporter: Nishan Shetty
>
> 1.Submit job
> 2.After successful execution of job before the Job history files are copied
> from intermediate done directory to done directory,NameNode got killed.
> 3.Restart the NameNode after mapreduce.jobhistory.move.interval-ms time is
> elapsed(default is 3 min).
> Observe that Job history files are not copied from intermediate done
> directory to done directory and also logs are not updated with any message
> Now submit another job observe that Job history files are not copied from
> intermediate done directory to done directory and also nothing is logged into
> historyserver logs.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
