hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Varun Saxena (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-5802) Application priority updates add pending apps to running ordering policy
Date Sun, 30 Oct 2016 08:01:58 GMT

    [ https://issues.apache.org/jira/browse/YARN-5802?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15619529#comment-15619529

Varun Saxena commented on YARN-5802:

Thanks [~bibinchundatt] for the patch. Changes overall LGTM.

Checkstyle and javadoc looks related. Can you fix them ?
Also in the test case added, we can explicitly set scheduler to capacity in config as test
makes sense only for capacity scheduler. instanceof check should not be required.
Moreover, shouldn't this test be added in TestCapacityScheduler instead fetching queues and
ordering policy in TestClientRMService ? Because although we found the issue via kill but
the real issue is during attempt removal from capacity scheduler.
However, I do not have a strong opinion on this. If its easy to add, probably add this test
in TestCapacityScheduler.

> Application priority updates add pending apps to running ordering policy
> ------------------------------------------------------------------------
>                 Key: YARN-5802
>                 URL: https://issues.apache.org/jira/browse/YARN-5802
>             Project: Hadoop YARN
>          Issue Type: Bug
>            Reporter: Bibin A Chundatt
>            Assignee: Bibin A Chundatt
>            Priority: Critical
>         Attachments: YARN-5802.0001.patch, YARN-5802.0002.patch
> {{LeafQueue#updateApplicationPriority}}
> {code}
>  getOrderingPolicy().removeSchedulableEntity(attempt);
>       // Update new priority in SchedulerApplication
>       attempt.setPriority(newAppPriority);
>       getOrderingPolicy().addSchedulableEntity(attempt);
> {code}
> We should add again to ordering policy only when  attempt available in first case.Else
during application attempt removal will try to iterate on killed application still available
in pending Ordering policy.Which can cause RM to crash.

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