[
https://issues.apache.org/jira/browse/YARN-6059?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15864539#comment-15864539
]
Konstantinos Karanasos commented on YARN-6059:
----------------------------------------------
Thanks Hitesh. Regarding your questions:
bq. We don't launch the paused container and instead we simply kill it.
bq. Hmm..since we kill the container an event for that will be raised
automatically. Am I missing something?
I understand that you immediately send a kill signal to the process, which is
good.
This is sufficient for killing the actual container and reclaiming the
resources in the NM, but it does not let the RM know what happened to the
container. Check, for example, the {{ContainerRelaunch}} or the
{{RecoveredContainerLaunch}} classes. Once the container has exited, they send
a {{ContainerExitEvent}}. We should send a similar event once the killing of
the paused container is done, setting the right exit code.
bq. In RecoveredContainerLaunch, indentation needs to be fixed.
My bad, that was a typo -- I meant {{ContainersLauncher}}: we need to fix the
indentation of the case statement.
> Update paused container state in the state store
> ------------------------------------------------
>
> Key: YARN-6059
> URL: https://issues.apache.org/jira/browse/YARN-6059
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Hitesh Sharma
> Assignee: Hitesh Sharma
> Attachments: YARN-5216-YARN-6059.001.patch,
> YARN-6059-YARN-5972.001.patch, YARN-6059-YARN-5972.002.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
