[ 
https://issues.apache.org/jira/browse/YARN-3508?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14608986#comment-14608986
 ] 

Wangda Tan commented on YARN-3508:
----------------------------------

bq. Ok, IIUC you want me to pass RMContext to 
ProportionalCapacityPreemptionPolicy and get event handler from there. Correct ?
Yes

bq. Any reason you want this changed?
RM components have a reference to RMContext is a convention, you can see most 
of RM components have RMContext. And this can avoid some future code changes if 
Proportional..Policy needs more information beyond dispatcher.

bq.  And anyways ProportionalCapacityPreemptionPolicy constructors are used 
only from tests.
Init is used by production code, I suggest to make constructor/init signature 
consistent.

> Preemption processing occuring on the main RM dispatcher
> --------------------------------------------------------
>
>                 Key: YARN-3508
>                 URL: https://issues.apache.org/jira/browse/YARN-3508
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: resourcemanager, scheduler
>    Affects Versions: 2.6.0
>            Reporter: Jason Lowe
>            Assignee: Varun Saxena
>         Attachments: YARN-3508.002.patch, YARN-3508.01.patch, 
> YARN-3508.03.patch, YARN-3508.04.patch
>
>
> We recently saw the RM for a large cluster lag far behind on the 
> AsyncDispacher event queue.  The AsyncDispatcher thread was consistently 
> blocked on the highly-contended CapacityScheduler lock trying to dispatch 
> preemption-related events for RMContainerPreemptEventDispatcher.  Preemption 
> processing should occur on the scheduler event dispatcher thread or a 
> separate thread to avoid delaying the processing of other events in the 
> primary dispatcher queue.



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

Reply via email to