[ https://issues.apache.org/jira/browse/YARN-5292?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15682104#comment-15682104 ]
Hitesh Sharma commented on YARN-5292: ------------------------------------- Thanks for the comments, [~arun suresh]. Regarding 1, the actual JIRA to avoid killing of opp. containers is [YARN-5216]. I'm working on a patch for that which works on top of the new schedule state. In our offline discussions we have talked about having APIs in ContainerManagementProtocol that allow PAUSE/RESUME on a container. The current implementation is only for opp. containers so there was no need to add anything to the ContainerManagementProtocol, but we can definitely extend it to guaranteed containers and make the required changes. I think the PAUSE/RESUME semantics are particularly of interest for Docker containers and I will be happy to help with any related work in this area. I have test cases as part of the patch for [YARN-5216] and that will test this code path. Please take a look at the patch a bit more closely so I can address other feedback. > Support for PAUSED container state > ---------------------------------- > > Key: YARN-5292 > URL: https://issues.apache.org/jira/browse/YARN-5292 > Project: Hadoop YARN > Issue Type: New Feature > Reporter: Hitesh Sharma > Assignee: Hitesh Sharma > Attachments: YARN-5292.001.patch, YARN-5292.002.patch, > YARN-5292.003.patch, yarn-5292.pdf > > > YARN-2877 introduced OPPORTUNISTIC containers, and YARN-5216 proposes to add > capability to customize how OPPORTUNISTIC containers get preempted. > In this JIRA we propose introducing a PAUSED container state. > When a running container gets preempted, it enters the PAUSED state, where it > remains until resources get freed up on the node then the preempted container > can resume to the running state. > > One scenario where this capability is useful is work preservation. How > preemption is done, and whether the container supports it, is implementation > specific. > For instance, if the container is a virtual machine, then preempt would pause > the VM and resume would restore it back to the running state. > If the container doesn't support preemption, then preempt would default to > killing the container. > -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org