[
https://issues.apache.org/jira/browse/YARN-2917?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14237557#comment-14237557
]
Rohith commented on YARN-2917:
------------------------------
Looking into deeply AsyncDispatcher, there is potential deadlock acros
{{AsyncDispatcher#dispatch()(System.exit)}} and
{{AsyncDispatcher#serviceStop()}}.
When ever dispatcher thread self exit on exception from the thread
{{eventHandlingThread}}, {{eventHandlingThread}} wait internally by jvm i.e
ApplicationShutdownHooks which runs ShutdownHook and join the shudownhook
thread. But in AsyncDispatcher#serviceStop(), eventHandlingThread is checked
for isAlive() which return always true result is while loop forever.
> RM get hanged if fail to store NodeLabels into store.
> -----------------------------------------------------
>
> Key: YARN-2917
> URL: https://issues.apache.org/jira/browse/YARN-2917
> Project: Hadoop YARN
> Issue Type: Bug
> Components: resourcemanager
> Reporter: Rohith
> Assignee: Rohith
> Priority: Critical
>
> I encoutered scenario where RM hanged while shutting down and keep on logging
> {{2014-12-03 19:32:44,283 INFO org.apache.hadoop.yarn.event.AsyncDispatcher:
> Waiting for AsyncDispatcher to drain.}}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
