hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Rohith Sharma K S (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-8679) [ATSv2] If HBase cluster is down, high chances that NM ContainerManager dispatcher get blocked
Date Fri, 17 Aug 2018 13:12:00 GMT

    [ https://issues.apache.org/jira/browse/YARN-8679?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16583923#comment-16583923
] 

Rohith Sharma K S commented on YARN-8679:
-----------------------------------------

Attaching a patch which relaxes a strict synchronization in PerNodeTimelineCollectorsAuxService
for the operations initializeContainer() and removeApplicationCollector(). 
[~sunilg] Can you please review the patch? 

> [ATSv2] If HBase cluster is down, high chances that NM ContainerManager dispatcher get
blocked
> ----------------------------------------------------------------------------------------------
>
>                 Key: YARN-8679
>                 URL: https://issues.apache.org/jira/browse/YARN-8679
>             Project: Hadoop YARN
>          Issue Type: Bug
>            Reporter: Rohith Sharma K S
>            Assignee: Rohith Sharma K S
>            Priority: Major
>         Attachments: YARN-8679.01.patch
>
>
> It is observed that if ATSv2 back end is down and client wait for few minutes, then NM
ContainerManager dispatcher thread is blocked. As a result, NM containers operations are stuck
in event processing.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org


Mime
View raw message