[ https://issues.apache.org/jira/browse/MAPREDUCE-4157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13261828#comment-13261828 ]
Ravi Prakash commented on MAPREDUCE-4157: ----------------------------------------- Hi Jason! I was thinking one of those parameters was the time the NM should wait before sending the SIGTERM. Alas, I was wrong ContainerLaunch.java:337 sends the SIGTERM pretty much instantly. That is messing me up in MAPREDUCE-4135. I'm still thinking having a small time delay before sending even the SIGTERM should be good. If we can do it on the NM and not on the RM, I think we might be able to conserve RM resources. > ResourceManager should not kill apps that are well behaved > ---------------------------------------------------------- > > Key: MAPREDUCE-4157 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4157 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: mrv2 > Affects Versions: 2.0.0 > Reporter: Jason Lowe > Assignee: Jason Lowe > Attachments: MAPREDUCE-4157.patch > > > Currently when the ApplicationMaster unregisters with the ResourceManager, > the RM kills (via the NMs) all the active containers for an application. > This introduces a race where the AM may be trying to clean up and may not > finish before it is killed. The RM should give the AM a chance to exit > cleanly on its own rather than always race with a pending kill on shutdown. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira