[
https://issues.apache.org/jira/browse/YARN-180?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13481872#comment-13481872
]
Vinod Kumar Vavilapalli commented on YARN-180:
----------------------------------------------
It looks like even after container assignment, there are some code paths (
atleast one right below the changes in the patch) where we don't return the
assigned container. One simpler approach to token generation is when
RMContainerImpl moves to ACQUIRED state. Thoughts?
> Capacity scheduler - containers that get reserved create container token to
> early
> ---------------------------------------------------------------------------------
>
> Key: YARN-180
> URL: https://issues.apache.org/jira/browse/YARN-180
> Project: Hadoop YARN
> Issue Type: Bug
> Components: capacityscheduler
> Affects Versions: 0.23.3
> Reporter: Thomas Graves
> Assignee: Arun C Murthy
> Priority: Critical
> Fix For: 2.0.3-alpha, 0.23.5
>
> Attachments: YARN-180.patch
>
>
> The capacity scheduler has the ability to 'reserve' containers.
> Unfortunately before it decides that it goes to reserved rather then
> assigned, the Container object is created which creates a container token
> that expires in roughly 10 minutes by default.
> This means that by the time the NM frees up enough space on that node for the
> container to move to assigned the container token may have expired.
--
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
