[
https://issues.apache.org/jira/browse/YARN-4761?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15179127#comment-15179127
]
Sangjin Lee commented on YARN-4761:
-----------------------------------
I'd like to discuss the unit test for this. I could essentially duplicate the
same test that was added to the {{TestCapacityScheduler}}. However, it might be
largely a copy-and-paste, and I'm not too happy about that but I could still do
that. Do let me know your thoughts on this.
A larger question is, we have a large amount of generic RM unit tests out
there, but they are exercised only against the capacity scheduler. Should we
try to find ways to exercise them against the fair scheduler as well? That
would be the most effective way of ensuring the soundness of any changes.
> NMs reconnecting with changed capabilities can lead to wrong cluster resource
> calculations on fair scheduler
> ------------------------------------------------------------------------------------------------------------
>
> Key: YARN-4761
> URL: https://issues.apache.org/jira/browse/YARN-4761
> Project: Hadoop YARN
> Issue Type: Bug
> Components: fairscheduler
> Affects Versions: 2.6.4
> Reporter: Sangjin Lee
> Assignee: Sangjin Lee
> Attachments: YARN-4761.01.patch
>
>
> YARN-3802 uncovered an issue with the scheduler where the resource
> calculation can be incorrect due to async event handling. It was subsequently
> fixed by YARN-4344, but it was never fixed for the fair scheduler.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
