[
https://issues.apache.org/jira/browse/MAPREDUCE-4437?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13415533#comment-13415533
]
Robert Joseph Evans commented on MAPREDUCE-4437:
------------------------------------------------
The change looks good, and it is generic enough to be sure that we recalculate
every time a container finishes.
+1.
> Race in MR ApplicationMaster can cause reducers to never be scheduled
> ---------------------------------------------------------------------
>
> Key: MAPREDUCE-4437
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-4437
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: applicationmaster, mrv2
> Affects Versions: 0.23.3, 2.1.0-alpha
> Reporter: Jason Lowe
> Assignee: Jason Lowe
> Priority: Critical
> Attachments: MAPREDUCE-4437.patch
>
>
> If the MR AM is notified of container completion by the RM before the AM
> receives notification of the container cleanup from the NM then it can fail
> to schedule reducers indefinitely. Logs showing the issue to follow.
--
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
