Wangda Tan commented on YARN-1631:

Just offline discussed with [~vinodkv] and [~jianhe], now our consensus is, if 
we allow skip first submitted application because of queueMax resource limit, 
it violates FIFO. This is different from user-limit, if we skip it, first 
application with large resource request may be starved. 

Another problem in your example is, app1 shouldn't reserve container in node_1 
because app_1's AM is running on that node too. Queue continuous reservation 
looking can solve this problem.

Closing it as won't fix. [~sunilg], please reopen it if you still think it's a 

> Container allocation issue in Leafqueue assignContainers()
> ----------------------------------------------------------
>                 Key: YARN-1631
>                 URL: https://issues.apache.org/jira/browse/YARN-1631
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: scheduler
>    Affects Versions: 2.2.0
>         Environment: SuSe 11 Linux 
>            Reporter: Sunil G
>            Assignee: Sunil G
>         Attachments: Yarn-1631.1.patch, Yarn-1631.2.patch
> Application1 has a demand of 8GB[Map Task Size as 8GB] which is more than 
> Node_1 can handle.
> Node_1 has a size of 8GB and 2GB is used by Application1's AM.
> Hence reservation happened for remaining 6GB in Node_1 by Application1.
> A new job is submitted with 2GB AM size and 2GB task size with only 2 Maps to 
> run.
> Node_2 also has 8GB capability.
> But Application2's AM cannot be launched in Node_2. And Application2 waits 
> longer as only 2 Nodes are available in cluster.

This message was sent by Atlassian JIRA

Reply via email to