[ 
https://issues.apache.org/jira/browse/YARN-9766?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16915533#comment-16915533
 ] 

Tarun Parimi commented on YARN-9766:
------------------------------------

[~eepayne], The change didn't affect any related unit tests in my local setup. 
So I am assuming it should be safe as {{old}} variable  is also not referred 
anywhere else in AbstractCSQueue.

Attaching the patch which removes the null check for old. Now 
CSQueueMetrics.forQueue() will be called every time even during queue refresh 
where {{old != null}} . But the cache in QueueMetrics, prevents creation of new 
metric instances and so should be fine. The test for QueueMetrics cache is 
properly covered in the existing TestQueueMetrics#testMetricsCache . 

Added a unit test in TestParentQueue which tests the scenario described in the 
jira.

Manually verified that queue metrics work fine in test cluster. Submitted 
sample DSShell jobs and verified basic Scheduling logic and capacity limits. 
Queue refresh also works fine.

cc [~sunilg],[~Prabhu Joseph]



> YARN CapacityScheduler QueueMetrics has missing metrics for parent queues 
> having same name
> ------------------------------------------------------------------------------------------
>
>                 Key: YARN-9766
>                 URL: https://issues.apache.org/jira/browse/YARN-9766
>             Project: Hadoop YARN
>          Issue Type: Bug
>    Affects Versions: 2.7.0
>            Reporter: Tarun Parimi
>            Assignee: Tarun Parimi
>            Priority: Major
>         Attachments: YARN-9766.001.patch
>
>
> In Capacity Scheduler, we enforce Leaf Queues to have unique names. But it is 
> not the case for Parent Queues. For example, we can have the below queue 
> hierarchy, where "b" is the queue name for two different queue paths root.a.b 
> and root.a.d.b . Since it is not a leaf queue this configuration works and 
> apps run fine in the leaf queues 'c'  and 'e'.
>  * root
>  ** a
>  *** b
>  **** c
>  *** d
>  **** b
>  ***** e
> But the jmx metrics does not show the metrics for the parent queue 
> "root.a.d.b" . We can see metrics only for "root.a.b" queue.
>  



--
This message was sent by Atlassian Jira
(v8.3.2#803003)

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

Reply via email to