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-4616) Default RM retry interval (30s) is too long
Date Thu, 21 Jan 2016 04:37:39 GMT

    [ https://issues.apache.org/jira/browse/YARN-4616?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15110037#comment-15110037
] 

Rohith Sharma K S commented on YARN-4616:
-----------------------------------------

It make sense to me, I am +1 for changing default value. 
Adding to above improvement, on second thought, why don't we achieve MTTR for RM services
less than a minute in several failures for all the client calls? It is just an thought , and
need to dig more about this how can it be achieved.

> Default RM retry interval (30s) is too long
> -------------------------------------------
>
>                 Key: YARN-4616
>                 URL: https://issues.apache.org/jira/browse/YARN-4616
>             Project: Hadoop YARN
>          Issue Type: Bug
>            Reporter: Jian He
>            Assignee: Jian He
>             Fix For: 2.8.0
>
>
> I think the default 30s for the RM retry interval is too long.
> The default node-heartbeat-interval is only 1s 



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

Mime
View raw message