[
https://issues.apache.org/jira/browse/YARN-1408?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14026159#comment-14026159
]
Vinod Kumar Vavilapalli commented on YARN-1408:
-----------------------------------------------
Pitching in late, apologies.
bq. After read code and thought about this problem, I think it's better to
store ResourceRequests deducted when allocate a container:
+1 for this proposal. It will add a transitory memory foot-print. Let's make
sure we benchmark this.
bq. 2. RMContainer can also store the AppSchedulingInfo object.
This is leaking abstractions and is a slippery slope. It is kind of weird to
have container talking to ASI. How about us adding a killContainer API in
scheduler (AbstractYarnScheduler) which checks the container state and adds
back the RR to ASI when needed? Killing a container is a generic functionality
anyways.
> Preemption caused Invalid State Event: ACQUIRED at KILLED and caused a task
> timeout for 30mins
> ----------------------------------------------------------------------------------------------
>
> Key: YARN-1408
> URL: https://issues.apache.org/jira/browse/YARN-1408
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: resourcemanager
> Affects Versions: 2.2.0
> Reporter: Sunil G
> Assignee: Sunil G
> Attachments: Yarn-1408.1.patch, Yarn-1408.2.patch, Yarn-1408.3.patch,
> Yarn-1408.4.patch, Yarn-1408.patch
>
>
> Capacity preemption is enabled as follows.
> * yarn.resourcemanager.scheduler.monitor.enable= true ,
> *
> yarn.resourcemanager.scheduler.monitor.policies=org.apache.hadoop.yarn.server.resourcemanager.monitor.capacity.ProportionalCapacityPreemptionPolicy
> Queue = a,b
> Capacity of Queue A = 80%
> Capacity of Queue B = 20%
> Step 1: Assign a big jobA on queue a which uses full cluster capacity
> Step 2: Submitted a jobB to queue b which would use less than 20% of cluster
> capacity
> JobA task which uses queue b capcity is been preempted and killed.
> This caused below problem:
> 1. New Container has got allocated for jobA in Queue A as per node update
> from an NM.
> 2. This container has been preempted immediately as per preemption.
> Here ACQUIRED at KILLED Invalid State exception came when the next AM
> heartbeat reached RM.
> ERROR
> org.apache.hadoop.yarn.server.resourcemanager.rmcontainer.RMContainerImpl:
> Can't handle this event at current state
> org.apache.hadoop.yarn.state.InvalidStateTransitonException: Invalid event:
> ACQUIRED at KILLED
> This also caused the Task to go for a timeout for 30minutes as this Container
> was already killed by preemption.
> attempt_1380289782418_0003_m_000000_0 Timed out after 1800 secs
--
This message was sent by Atlassian JIRA
(v6.2#6252)
