[ 
https://issues.apache.org/jira/browse/YARN-4000?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14744884#comment-14744884
 ] 

Jian He commented on YARN-4000:
-------------------------------

[~varun_saxena], thanks for working on the patch, some comments:
- We may not need a new RMAppKillEvent; we can add a new diagnostics field into 
the existing RMAppEvent, which will be useful for other types of events too.
- the appDiagnosticsBeforeKilling in RMAppImpl is also not needed, we can reuse 
the diagnostics object. 
- CapacityScheduler#addApplication is now getting a bit complex, could you 
separate a new method called recoverApplication and put the recovery logic in 
there ?



> RM crashes with NPE if leaf queue becomes parent queue during restart
> ---------------------------------------------------------------------
>
>                 Key: YARN-4000
>                 URL: https://issues.apache.org/jira/browse/YARN-4000
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: capacityscheduler, resourcemanager
>    Affects Versions: 2.6.0
>            Reporter: Jason Lowe
>            Assignee: Varun Saxena
>         Attachments: YARN-4000.01.patch, YARN-4000.02.patch
>
>
> This is a similar situation to YARN-2308.  If an application is active in 
> queue A and then the RM restarts with a changed capacity scheduler 
> configuration where queue A becomes a parent queue to other subqueues then 
> the RM will crash with a NullPointerException.



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

Reply via email to