[
https://issues.apache.org/jira/browse/YARN-8095?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16440283#comment-16440283
]
kyungwan nam commented on YARN-8095:
------------------------------------
I attached a patch, which respect the accessible-node-lables even though
non-exclusive allocation.
when the capacity-scheduler configs are as follows
{code:java}
yarn.scheduler.capacity.root.batch.accessible-node-labels=label1
yarn.scheduler.capacity.root.longlived.accessible-node-labels= {code}
I’ve seen that apps submitted to root.longlived does not allocated on ‘label1’
partition.
> Allow disable non-exclusive allocation
> --------------------------------------
>
> Key: YARN-8095
> URL: https://issues.apache.org/jira/browse/YARN-8095
> Project: Hadoop YARN
> Issue Type: Improvement
> Components: capacity scheduler
> Affects Versions: 2.8.3
> Reporter: kyungwan nam
> Priority: Major
> Attachments: YARN-8095-branch-2.8.001.patch
>
>
> We have 'longlived' Queue, which is used for long-lived apps.
> In situation where default Partition resources are not enough, containers for
> long-lived app can be allocated to sharable Partition.
> Since then, containers for long-lived app can be easily preempted.
> We don’t want long-lived apps to be killed abruptly.
> Currently, non-exclusive allocation can happen regardless of whether the
> queue is accessible to the sharable Partition.
> It would be good if non-exclusive allocation can be disabled at queue level.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
