[
https://issues.apache.org/jira/browse/YARN-1049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13740265#comment-13740265
]
Zhijie Shen commented on YARN-1049:
-----------------------------------
bq. ContainerExitStatus defines a few constant with special exit status values
(0,-1000, -100, -101). This is incorrect, we should not define any special
constants and limit to return the actual process exist status code.
In addition to ContainerExitStatus, ExitCode also defines 137 and 143. However,
except these values, a container's exit code usually comes from the exit value
of its process. Are you concerned that the value from the process may conflict
the self defined one?
> ContainerExistStatus and ContainerState are defined incorrectly
> ---------------------------------------------------------------
>
> Key: YARN-1049
> URL: https://issues.apache.org/jira/browse/YARN-1049
> Project: Hadoop YARN
> Issue Type: Bug
> Components: api
> Affects Versions: 2.1.0-beta
> Reporter: Alejandro Abdelnur
> Priority: Blocker
> Fix For: 2.1.0-beta
>
>
> ContainerExitStatus defines a few constant with special exit status values
> (0,-1000, -100, -101). This is incorrect, we should not define any special
> constants and limit to return the actual process exist status code.
> ContainerState should include PREEMPTED (when preempted by YARN), LOST (when
> the NM crashes).
> With the current behavior is impossible to determine if a container has been
> preempted or lost due to a NM crash.
> Marking it as a blocker for 2.1.0 as this is an API/behavior change.
--
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
