[ 
https://issues.apache.org/jira/browse/YARN-2005?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14296936#comment-14296936
 ] 

Sunil G commented on YARN-2005:
-------------------------------

bq.  I've seen use a different app name each time they submit
Yes. This is a real valid point in a production cluster. I hope the user will 
be same for these, hence user can play a bigger role in identifying such 
problematic apps.
as mentioned in point 2, 
* failed app can be restricted to run in same node on a reattempt scenario
* If an app is failed for a given user, and same user submitting another 
application, its good to scheduler that also in a different node. 

bq.but the biggest problem we're seeing probably doesn't need anything that 
fancy to solve 80% of the cases we see.
I agree that a simple logic can be shaped up first, and can see the 
feasibility. Then an analysis on how much it can really help. After that its 
better to go for complexity. Kindly suggests your thoughts on same.

> Blacklisting support for scheduling AMs
> ---------------------------------------
>
>                 Key: YARN-2005
>                 URL: https://issues.apache.org/jira/browse/YARN-2005
>             Project: Hadoop YARN
>          Issue Type: Improvement
>          Components: resourcemanager
>    Affects Versions: 0.23.10, 2.4.0
>            Reporter: Jason Lowe
>
> It would be nice if the RM supported blacklisting a node for an AM launch 
> after the same node fails a configurable number of AM attempts.  This would 
> be similar to the blacklisting support for scheduling task attempts in the 
> MapReduce AM but for scheduling AM attempts on the RM side.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to