Sunil G updated YARN-2003:
    Attachment: 0024-YARN-2003.patch

Hi [~leftnoteasy]
I have done some more analysis on test failure. As per the YARN-3533, fix is 
done in {{launchAM}} of MockRM. I was using MockRM apis in below way earlier.
RMApp app1 = rm.submitApp(1 * GB, appPriority1);
MockAM am1 = rm.sendAMLaunched(app1.getCurrentAppAttempt()

{{launchAM}} already wraps nm.nodeHeartbeat and sendAMLaunched. And after 
YARN-3533, its assured that attempt state is SCHEDULED before calling 
nodeHeartbeat. If we directly use as per above code segment, its some times 
possible that before app attempt is SCHEDULED, nm heartbeats arrives and 
attempt may be still ALLOCATED.

Hence I changed my test cases to use this wrapped api {{launchAM}}  from 
MockRM. Uploading a patch with same.

> Support for Application priority : Changes in RM and Capacity Scheduler
> -----------------------------------------------------------------------
>                 Key: YARN-2003
>                 URL: https://issues.apache.org/jira/browse/YARN-2003
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: resourcemanager
>            Reporter: Sunil G
>            Assignee: Sunil G
>         Attachments: 0001-YARN-2003.patch, 00010-YARN-2003.patch, 
> 0002-YARN-2003.patch, 0003-YARN-2003.patch, 0004-YARN-2003.patch, 
> 0005-YARN-2003.patch, 0006-YARN-2003.patch, 0007-YARN-2003.patch, 
> 0008-YARN-2003.patch, 0009-YARN-2003.patch, 0011-YARN-2003.patch, 
> 0012-YARN-2003.patch, 0013-YARN-2003.patch, 0014-YARN-2003.patch, 
> 0015-YARN-2003.patch, 0016-YARN-2003.patch, 0017-YARN-2003.patch, 
> 0018-YARN-2003.patch, 0019-YARN-2003.patch, 0020-YARN-2003.patch, 
> 0021-YARN-2003.patch, 0022-YARN-2003.patch, 0023-YARN-2003.patch, 
> 0024-YARN-2003.patch
> AppAttemptAddedSchedulerEvent should be able to receive the Job Priority from 
> Submission Context and store.
> Later this can be used by Scheduler.

This message was sent by Atlassian JIRA

Reply via email to