hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jason Lowe (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-5120) Metric for RM async dispatcher queue size
Date Mon, 23 May 2016 17:40:12 GMT

    [ https://issues.apache.org/jira/browse/YARN-5120?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15296700#comment-15296700
] 

Jason Lowe commented on YARN-5120:
----------------------------------

Note there are multiple important dispatchers in the ResourceManager for determining health.
 The main AsyncDispatcher is one, but the SchedulerEventDispatcher is another.  We've seen
instances in the past where the primary dispatcher was fine and the backlog was all on the
scheduler dispatcher queue.

> Metric for RM async dispatcher queue size
> -----------------------------------------
>
>                 Key: YARN-5120
>                 URL: https://issues.apache.org/jira/browse/YARN-5120
>             Project: Hadoop YARN
>          Issue Type: Improvement
>          Components: resourcemanager
>            Reporter: Giovanni Matteo Fumarola
>            Priority: Minor
>
> It is difficult to identify the health of the RM AsyncDispatcher. 
> Solution: Add a metric for the AsyncDispatcher queue size. 



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

---------------------------------------------------------------------
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