[ 
https://issues.apache.org/jira/browse/MAPREDUCE-3496?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13161020#comment-13161020
 ] 

Jonathan Eagles commented on MAPREDUCE-3496:
--------------------------------------------

The solution could have been solve on the client side or the server side. This 
patch implements a client side solution. This will ensure predictable results 
no matter what scheduler is being used. Furthermore, the server container 
solution is not compromised (for example switching server to using 
LinkedHashMap from HashMap) nor relied upon to satisfy client view requirements.
                
> Yarn initializes ACL operations from capacity scheduler config in a 
> non-deterministic order
> -------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-3496
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-3496
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: mrv2
>    Affects Versions: 0.23.1, 0.24.0
>            Reporter: Jonathan Eagles
>            Assignee: Jonathan Eagles
>         Attachments: MAPREDUCE-3496.patch
>
>
> 'mapred queue -showacls' does not output put acls in a predictable manner. 
> This is a regression from previous versions.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Reply via email to