hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Arun Suresh (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-6251) Fix Scheduler locking issue introduced by YARN-6216
Date Fri, 18 Aug 2017 21:55:00 GMT

    [ https://issues.apache.org/jira/browse/YARN-6251?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16133673#comment-16133673

Arun Suresh commented on YARN-6251:

Thanks for taking a look [~leftnoteasy],

bq. However, if in the code path, we expect resource released before new resource allocated
(like continuous-reservation-looking), otherwise cluster resource will be overflowed temporally.
Yup, but you probably wont see an overflow - but, yes the scheduler might see a delay in being
able to consume the released resourced (during container decrease and demote) until after
the event is processed. Which I think it might be tolerable given that the alternative can
be a deadlock :) That is why I had to modify some of the test cases to drain the dispatcher
queue before asserting some of the conditions.

Will update the patch with your suggestions. Thanks 

> Fix Scheduler locking issue introduced by YARN-6216
> ---------------------------------------------------
>                 Key: YARN-6251
>                 URL: https://issues.apache.org/jira/browse/YARN-6251
>             Project: Hadoop YARN
>          Issue Type: Bug
>            Reporter: Arun Suresh
>            Assignee: Arun Suresh
>         Attachments: YARN-6251.001.patch, YARN-6251.002.patch, YARN-6251.003.patch, YARN-6251.004.patch
> Opening to track a locking issue that was uncovered when running a custom SLS AMSimulator.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org

View raw message