[
https://issues.apache.org/jira/browse/MAPREDUCE-3084?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13122857#comment-13122857
]
Vinod Kumar Vavilapalli commented on MAPREDUCE-3084:
----------------------------------------------------
I discussed with Sid regarding this and found that there is bigger race here.
ContainerLaunch does a bunch of setup steps needed for the container and if a
kill arrives during any of it, the container will never be killed. We do cancel
the Future, but that is without any interrupts. Also we try to kill the
process, but the PID may not be available at that point of time.
This part of the code needs a minor redesign.
> race when KILL_CONTAINER is received for a LOCALIZED container
> --------------------------------------------------------------
>
> Key: MAPREDUCE-3084
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-3084
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: mrv2
> Affects Versions: 0.23.0
> Reporter: Siddharth Seth
> Assignee: Hitesh Shah
> Priority: Blocker
> Attachments: MR-3084.1.patch, MR-3084.wip.patch
>
>
> Depending on when ContainersLaunch starts a container, {{KILL_CONTAINER}}
> when container state is {{LOCALIZED}} ({{LAUNCH_CONTAINER}} event already
> sent) can end up generating a {{CONTAINER_LAUNCHED}} event - which isn't
> handled by ContainerState: {{KILLING}}. Also, the launched container won't be
> killed since {{CLEANUP_CONTAINER}} would have already been processed.
--
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
