[jira] [Commented] (YARN-2306) leak of reservation metrics (fair scheduler)

[Tsuyoshi OZAWA (JIRA)]

    [ 
https://issues.apache.org/jira/browse/YARN-2306?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14119707#comment-14119707
 ] 

Tsuyoshi OZAWA commented on YARN-2306:
--------------------------------------

Thanks for your updating, [~zhiguohong].
The test cases you added looks good to me. I know it's not related to your 
patch directory and minor nits, but how about stopping resourceManager in 
tearDown method to fix the test failures like this?

{code}
  @After
  public void tearDown() {
    scheduler = null;
    if (resourceManager != null) {
      resourceManager.stop();
    }
    QueueMetrics.clearQueueMetrics();
    DefaultMetricsSystem.shutdown();
  }
{code}

> leak of reservation metrics (fair scheduler)
> --------------------------------------------
>
>                 Key: YARN-2306
>                 URL: https://issues.apache.org/jira/browse/YARN-2306
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: scheduler
>            Reporter: Hong Zhiguo
>            Assignee: Hong Zhiguo
>            Priority: Minor
>         Attachments: YARN-2306-2.patch, YARN-2306.patch
>
>
> This only applies to fair scheduler. Capacity scheduler is OK.
> When appAttempt or node is removed, the metrics for 
> reservation(reservedContainers, reservedMB, reservedVCores) is not reduced 
> back.
> These are important metrics for administrator. The wrong metrics confuses may 
> confuse them. 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)