hadoop-mapreduce-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Robert Joseph Evans (JIRA)" <j...@apache.org>
Subject [jira] [Resolved] (MAPREDUCE-4480) T_ATTEMPT_KILLED after SUCCEEDED can happen for reduces too
Date Thu, 15 Nov 2012 16:06:13 GMT

     [ https://issues.apache.org/jira/browse/MAPREDUCE-4480?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Robert Joseph Evans resolved MAPREDUCE-4480.
--------------------------------------------

    Resolution: Not A Problem

This appears to no longer be a problem.  Several JIRA have modified the MapRetoractiveKillTransition
to the point that now the only time that this can happen is if a successful reduce task attempt
is killed after it has succeeded, which should never happen, because the reduce succeeded.
                
> T_ATTEMPT_KILLED after SUCCEEDED can happen for reduces too 
> ------------------------------------------------------------
>
>                 Key: MAPREDUCE-4480
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4480
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>    Affects Versions: 3.0.0, 2.0.2-alpha
>            Reporter: Robert Joseph Evans
>            Priority: Critical
>
> This does not seem to impact 0.23.  If speculative execution is enabled then a T_ATTEMPT_KILLED
event can come in after the task has transitioned to SUCCEEDED.  This causes the MapRetroactiveKilledTransition
to kill the Job, because it expects to only handle map tasks.

--
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

Mime
View raw message