[
https://issues.apache.org/jira/browse/YARN-6956?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16117772#comment-16117772
]
Karthik Kambatla commented on YARN-6956:
----------------------------------------
There are a series of follow-up changes to be made:
# Prioritize the RR to check: at a particular priority level, maybe we should
look at the RR with least resources.
# Consider more than one RR. While considering all RRs might be too expensive,
we could consider a configurable number of RRs.
# Honor delay scheduling for preemption as well. If we are unable to identify
containers to preempt on a node and there is a corresponding rack-local RR,
consider finding any node on the rack. Similarly, relax to the entire cluster.
[~Steven Rand], do these suggestions make sense? Will you be interested in
contributing?
> preemption may only consider resource requests for one node
> -----------------------------------------------------------
>
> Key: YARN-6956
> URL: https://issues.apache.org/jira/browse/YARN-6956
> Project: Hadoop YARN
> Issue Type: Bug
> Components: fairscheduler
> Affects Versions: 2.9.0, 3.0.0-beta1
> Environment: CDH 5.11.0
> Reporter: Steven Rand
>
> I'm observing the following series of events on a CDH 5.11.0 cluster, which
> seem to be possible after YARN-6163:
> 1. An application is considered to be starved, so {{FSPreemptionThread}}
> calls {{identifyContainersToPreempt}}, and that calls
> {{FSAppAttempt#getStarvedResourceRequests}} to get a list of
> {{ResourceRequest}} instances that are enough to address the app's starvation.
> 2. The first {{ResourceRequest}} that {{getStarvedResourceRequests}} sees is
> enough to address the app's starvation, so we break out of the loop over
> {{appSchedulingInfo.getAllResourceRequests()}} after only one iteration:
> https://github.com/apache/hadoop/blob/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSAppAttempt.java#L1180.
> We return only this one {{ResourceRequest}} back to the
> {{identifyContainersToPreempt}} method.
> 3. It turns out that this particular {{ResourceRequest}} happens to have a
> value for {{getResourceName}} that identifies a specific node in the cluster.
> This causes preemption to only consider containers on that node, and not the
> rest of the cluster.
> [~kasha], does that make sense? I'm happy to submit a patch if I'm
> understanding the problem correctly.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
