Jason Lowe commented on YARN-4280:

I also thought originally that relaxing the queue max cap check could resolve 
the problem, but that can also create new ones besides the reserved+allocated > 
total issue.  The problem is that reservations are intended to become actual 
containers.  If the resources to satisfy the reservation come from the queue 
where the reservation took place then we're fine.  However if the resources to 
satisfy the reservation came from a peer queue then suddenly we're running the 
queue over max capacity and violating constraints even without any outstanding 
reservations at that point.

> CapacityScheduler reservations may not prevent indefinite postponement on a 
> busy cluster
> ----------------------------------------------------------------------------------------
>                 Key: YARN-4280
>                 URL: https://issues.apache.org/jira/browse/YARN-4280
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: capacity scheduler
>    Affects Versions: 2.6.1, 2.8.0, 2.7.1
>            Reporter: Kuhu Shukla
>            Assignee: Kuhu Shukla
> Consider the following scenario:
> There are 2 queues A(25% of the total capacity) and B(75%), both can run at 
> total cluster capacity. There are 2 applications, appX that runs on Queue A, 
> always asking for 1G containers(non-AM) and appY runs on Queue B asking for 2 
> GB containers.
> The user limit is high enough for the application to reach 100% of the 
> cluster resource. 
> appX is running at total cluster capacity, full with 1G containers releasing 
> only one container at a time. appY comes in with a request of 2GB container 
> but only 1 GB is free. Ideally, since appY is in the underserved queue, it 
> has higher priority and should reserve for its 2 GB request. Since this 
> request puts the alloc+reserve above total capacity of the cluster, 
> reservation is not made. appX comes in with a 1GB request and since 1GB is 
> still available, the request is allocated. 
> This can continue indefinitely causing priority inversion.

This message was sent by Atlassian JIRA

Reply via email to