[
https://issues.apache.org/jira/browse/YARN-8513?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16584787#comment-16584787
]
Chen Yufei commented on YARN-8513:
----------------------------------
New jstack/top and RM logs are uploaded and prefixed with yarn3. We upgraded to
Hadoop 3.1.1 yesterday and encounter this problem several times.
The problem seems reproducible when one queue is near fully utilized. Killing
current active RM can not solve the problem. We have to kill some jobs in the
fully utilized queue in order to submit new jobs.
Debug log shows that CapacityScheduler repeatedly trying to schedule on a
specific node, but as queue resource has exceeded resource limit, allocation
proposal won't be accepted. top command shows only one thread with near 100%
CPU usage, strace shows this thread is the one trying to do the allocation and
flushing out logs.
I've tried to dig into source code, but can't find out why RM repeatedly trying
to schedule on a specific node.
Some notes about our setup:
* 3 partitions: default, sim, gpu
* 4 queues: dev & mkt (10% capacity, max 90%), dev-daily & mkt-daily (40%
capacity, max 100%)
* preemption disabled
> CapacityScheduler infinite loop when queue is near fully utilized
> -----------------------------------------------------------------
>
> Key: YARN-8513
> URL: https://issues.apache.org/jira/browse/YARN-8513
> Project: Hadoop YARN
> Issue Type: Bug
> Components: capacity scheduler, yarn
> Affects Versions: 3.1.0, 2.9.1
> Environment: Ubuntu 14.04.5 and 16.04.4
> YARN is configured with one label and 5 queues.
> Reporter: Chen Yufei
> Priority: Major
> Attachments: jstack-1.log, jstack-2.log, jstack-3.log, jstack-4.log,
> jstack-5.log, top-during-lock.log, top-when-normal.log, yarn3-jstack1.log,
> yarn3-jstack2.log, yarn3-jstack3.log, yarn3-jstack4.log, yarn3-jstack5.log,
> yarn3-resourcemanager.log, yarn3-top
>
>
> ResourceManager does not respond to any request when queue is near fully
> utilized sometimes. Sending SIGTERM won't stop RM, only SIGKILL can. After RM
> restart, it can recover running jobs and start accepting new ones.
>
> Seems like CapacityScheduler is in an infinite loop printing out the
> following log messages (more than 25,000 lines in a second):
>
> {{2018-07-10 17:16:29,227 INFO
> org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.ParentQueue:
> assignedContainer queue=root usedCapacity=0.99816763
> absoluteUsedCapacity=0.99816763 used=<memory:16170624, vCores:1577>
> cluster=<memory:29441544, vCores:5792>}}
> {{2018-07-10 17:16:29,227 INFO
> org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacityScheduler:
> Failed to accept allocation proposal}}
> {{2018-07-10 17:16:29,227 INFO
> org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.allocator.AbstractContainerAllocator:
> assignedContainer application attempt=appattempt_1530619767030_1652_000001
> container=null
> queue=org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.allocator.RegularContainerAllocator@14420943
> clusterResource=<memory:29441544, vCores:5792> type=NODE_LOCAL
> requestedPartition=}}
>
> I encounter this problem several times after upgrading to YARN 2.9.1, while
> the same configuration works fine under version 2.7.3.
>
> YARN-4477 is an infinite loop bug in FairScheduler, not sure if this is a
> similar problem.
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
