hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Amareshwari Sriramadasu (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-3734) Force dirhandler.checkDirs() in ResourceLocalizationService when there is a localization failure
Date Fri, 12 Jun 2015 06:10:01 GMT

    [ https://issues.apache.org/jira/browse/YARN-3734?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14583005#comment-14583005
] 

Amareshwari Sriramadasu commented on YARN-3734:
-----------------------------------------------

[~lavkesh], Can you look at processing of checkdirs()? How much time can that take? Is it
thread safe?

> Force dirhandler.checkDirs() in ResourceLocalizationService when there is a localization
failure
> ------------------------------------------------------------------------------------------------
>
>                 Key: YARN-3734
>                 URL: https://issues.apache.org/jira/browse/YARN-3734
>             Project: Hadoop YARN
>          Issue Type: Bug
>            Reporter: Lavkesh Lahngir
>            Assignee: Lavkesh Lahngir
>
> This scenario happens when a disk goes bad in between node health checks. In that case
all containers will be failing. 
> Would that be okay to force a checkdirs() when there is an IOException while localizing
a resource so that next container will not fail due to disk issues?



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

Mime
View raw message