[
https://issues.apache.org/jira/browse/YARN-10486?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17232736#comment-17232736
]
Benjamin Teke commented on YARN-10486:
--------------------------------------
Hi [~pbacsko],
Thanks for the patches/approaches. While the first solution, the approach of
using a fixup queue is less risky as it doesn't modify existing behaviour, I
think it is a bit "hacky" and because the fixup queue will be in the
configuration it is less user friendly (if the user want to add a new queue
they will need to edit the "fixup" queue's capacity as well, which can lead to
strange setups). That's why I generally prefer the second option, let's give an
option to ease the sum of capacity constraint.
Regarding the code itself I had a few minor nits:
* in patch 002:
** there is a typo in the String literal in
{{FSConfigToCSConfigRuleHandler.java }}*fixuo*QueueNeeded.action
** I think it would be beneficial to check the autogenerated queue's status
(stopped)
* in patch 004:
** WeightToCapacityConversionUtil still has the fixup queue in the javadoc
> FS-CS converter: handle case when weight=0
> ------------------------------------------
>
> Key: YARN-10486
> URL: https://issues.apache.org/jira/browse/YARN-10486
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: yarn
> Reporter: Peter Bacsko
> Assignee: Peter Bacsko
> Priority: Major
> Labels: fs2cs
> Attachments: YARN-10486-001.patch, YARN-10486-002.patch,
> YARN-10486-003-approach2.patch, YARN-10486-004-approach2.patch
>
>
> We can encounter an ArithmeticException if there is a single or multiple
> queues under a parent with zero weight.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
