[jira] [Commented] (YARN-4931) Preempted resources go back to the same application

["Miles Crawford (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/YARN-4931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15235591#comment-15235591 ] 

Miles Crawford commented on YARN-4931:
--------------------------------------

If it is helpful, it looks like these applications have a huge number of outstanding container requests (thousands!) and this number has to get drained down to zero before other applications requesting only 1 container can be served.

> Preempted resources go back to the same application
> ---------------------------------------------------
>
>                 Key: YARN-4931
>                 URL: https://issues.apache.org/jira/browse/YARN-4931
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: fairscheduler
>    Affects Versions: 2.7.2
>            Reporter: Miles Crawford
>         Attachments: resourcemanager.log
>
>
> Sometimes a queue that needs resources causes preemption - but the preempted containers are just allocated right back to the application that just released them!
> Here is a tiny application (0007) that wants resources, and a container is preempted from application 0002 to satisfy it:
> {code}
> 2016-04-07 21:08:13,463 INFO org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FairScheduler (FairSchedulerUpdateThread): Should preempt <memory:448, vCores:0> res for queue root.default: resDueToMinShare = <memory:0, vCores:0>, resDueToFairShare = <memory:448, vCores:0>
> 2016-04-07 21:08:13,463 INFO org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FairScheduler (FairSchedulerUpdateThread): Preempting container (prio=1res=<memory:15264, vCores:1>) from queue root.milesc
> 2016-04-07 21:08:13,463 INFO org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptMetrics (FairSchedulerUpdateThread): Non-AM container preempted, current appAttemptId=appattempt_1460047303577_0002_000001, containerId=container_1460047303577_0002_01_001038, resource=<memory:15264, vCores:1>
> 2016-04-07 21:08:13,463 INFO org.apache.hadoop.yarn.server.resourcemanager.rmcontainer.RMContainerImpl (FairSchedulerUpdateThread): container_1460047303577_0002_01_001038 Container Transitioned from RUNNING to KILLED
> {code}
> But then a moment later, application 00002 gets the container right back:
> {code}
> 2016-04-07 21:08:13,844 INFO org.apache.hadoop.yarn.server.resourcemanager.scheduler.SchedulerNode (ResourceManager Event Processor): Assigned container container_1460047303577_0002_01_001039 of capacity <memory:15264, vCores:1> on host ip-10-12-40-63.us-west-2.compute.internal:8041, which has 13 containers, <memory:241248, vCores:18> used and <memory:416, vCores:46> available after allocation
> 2016-04-07 21:08:14,555 INFO org.apache.hadoop.yarn.server.resourcemanager.rmcontainer.RMContainerImpl (IPC Server handler 59 on 8030): container_1460047303577_0002_01_001039 Container Transitioned from ALLOCATED to ACQUIRED
> 2016-04-07 21:08:14,845 INFO org.apache.hadoop.yarn.server.resourcemanager.rmcontainer.RMContainerImpl (ResourceManager Event Processor): container_1460047303577_0002_01_001039 Container Transitioned from ACQUIRED to RUNNING
> {code}
> This results in new applications being unable to even get an AM, and never starting at all.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)