hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Bikas Saha (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-498) Unmanaged AM launcher does not set various constants in env for an AM, also does not handle failed AMs properly
Date Fri, 22 Mar 2013 18:05:16 GMT

    [ https://issues.apache.org/jira/browse/YARN-498?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13611000#comment-13611000
] 

Bikas Saha commented on YARN-498:
---------------------------------

I had originally left the loop without a timeout because eventually YARN is supposed to recognize
that the app succeeded/killed/failed or it would expire and fail the app. Is that not happening?

I agree that a timeout would be more useful. How about adding a timeout to the monitoring
method such that it throws an exception if the desired end state is not reached within that
time. With the code change in the patch, it would end up crashing with an NPE, right?
                
> Unmanaged AM launcher does not set various constants in env for an AM, also does not
handle failed AMs properly
> ---------------------------------------------------------------------------------------------------------------
>
>                 Key: YARN-498
>                 URL: https://issues.apache.org/jira/browse/YARN-498
>             Project: Hadoop YARN
>          Issue Type: Bug
>            Reporter: Hitesh Shah
>            Assignee: Hitesh Shah
>         Attachments: YARN-498.1.patch, YARN-498.2.patch, YARN-498.wip.patch
>
>
> Currently, it only sets the app attempt id which is really not required as AMs are only
expected to extract it from the container id.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message