hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sandy Ryza (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-352) Inconsistent picture of how a container was killed when querying RM and NM in case of preemption
Date Tue, 22 Jan 2013 23:48:12 GMT

    [ https://issues.apache.org/jira/browse/YARN-352?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13560192#comment-13560192
] 

Sandy Ryza commented on YARN-352:
---------------------------------

Perhaps the exit code should not be overloaded to contain this kind of information, and a
ContainerStatus should contain a separate enum to report on why the container was killed,
as opposed to what it returned when it died?
                
> Inconsistent picture of how a container was killed when querying RM and NM in case of
preemption
> ------------------------------------------------------------------------------------------------
>
>                 Key: YARN-352
>                 URL: https://issues.apache.org/jira/browse/YARN-352
>             Project: Hadoop YARN
>          Issue Type: Bug
>            Reporter: Hitesh Shah
>
> When the RM preempts a container, it records the exit status as -100. However, at the
NM, it registers the preempted container's exit status as simply killed by an external via
SIGTERM or SIGKILL.
> When the AM queries the RM and NM for the same container's status, it will get 2 different
values. 
> When killing a container, the exit reason should likely be more defined via an exit status
code for the AM to act on in addition to providing of the diagnostic messages that can contain
more detailed information ( though probably not programmatically interpret-able by the AM
). 

--
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