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

Posted to mapreduce-issues@hadoop.apache.org by "Ravi Prakash (JIRA)" <ji...@apache.org> on 2013/07/15 20:16:58 UTC

[jira] [Moved] (MAPREDUCE-5390) JOB_FAIL and JOB_KILL have different behaviors when they should ideally be same / similar.

     [ https://issues.apache.org/jira/browse/MAPREDUCE-5390?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ravi Prakash moved HADOOP-9732 to MAPREDUCE-5390:
-------------------------------------------------

    Affects Version/s:     (was: 2.2.0)
                           (was: 0.23.9)
                           (was: 3.0.0)
                       2.2.0
                       3.0.0
                       0.23.9
                  Key: MAPREDUCE-5390  (was: HADOOP-9732)
              Project: Hadoop Map/Reduce  (was: Hadoop Common)
    
> JOB_FAIL and JOB_KILL have different behaviors when they should ideally be same / similar.
> ------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-5390
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-5390
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>    Affects Versions: 0.23.9, 3.0.0, 2.2.0
>            Reporter: Ravi Prakash
>
> After MAPREDUCE-5317, both JOB_KILL and JOB_FAIL wait for all the tasks to die / be killed, and then move to their final states. We can make these two code paths the same. Ideally KILL_WAIT should also have a timeout like the one MAPREDUCE-5317 introduces.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira