[
https://issues.apache.org/jira/browse/YARN-4934?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15233383#comment-15233383
]
Sunil G commented on YARN-4934:
-------------------------------
After analysis I found that this issue is not related to the fix made by this
patch or earlier rather its a timing issue. YARN-4890 is tracking this problem
and a patch is shared there.
Other webservices failures are due to "Address already in use" pblm. YARN-4909
is tracking this sporadic issue. Other issues are know pblms due to DNS.
> Reserved Resource for QueueMetrics needs to be handled correctly in few cases
> ------------------------------------------------------------------------------
>
> Key: YARN-4934
> URL: https://issues.apache.org/jira/browse/YARN-4934
> Project: Hadoop YARN
> Issue Type: Bug
> Components: capacity scheduler
> Affects Versions: 2.9.0
> Reporter: Sunil G
> Assignee: Sunil G
> Attachments: 0001-YARN-4934.patch
>
>
> Reseved Resource for QueueMetrics needs to be decremented correctly in cases
> like below:
> - when a reserved container is allocated
> - when node is lost/ disconnected.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
