[
https://issues.apache.org/jira/browse/TEZ-3128?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15157279#comment-15157279
]
Hitesh Shah commented on TEZ-3128:
----------------------------------
[~ozawa] I dont think the delayed container manager thread is the issue here.
[~sseth] can you add more details/logs on this.
I see the following as per code:
- dagappmaster shuts down yarn scheduler service but it does not kill
containers on shutdown - just releases them via amrmclient
- TezTaskCommunicatorImpl on stop() does nothing to kill containers.
It seems like the container launcher is the one trying shut down containers for
some reason. Maybe we should just release containers via the scheduler service
instead of trying to stop them?
> Avoid stopping containers on the AM shutdown thread
> ---------------------------------------------------
>
> Key: TEZ-3128
> URL: https://issues.apache.org/jira/browse/TEZ-3128
> Project: Apache Tez
> Issue Type: Bug
> Affects Versions: 0.8.0-alpha
> Reporter: Siddharth Seth
> Assignee: Tsuyoshi Ozawa
> Labels: newbie
> Attachments: TEZ-3128.001.patch
>
>
> During an AM shutdown, the TaskCommunicator is also shutdown and it tries to
> stop containers in the shutdown thread itself. This can cause the AM shutdown
> to block if NMs are not available.
> This likely affects 0.7 as well.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
