[
https://issues.apache.org/jira/browse/YARN-2005?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14296953#comment-14296953
]
Jason Lowe commented on YARN-2005:
----------------------------------
As I mentioned earlier, as a first step I think we could implement an
app-specific blacklisting approach similar to what is done by the MapReduce AM
today. We would track, per application, the nodes that have failed an AM
attempt and refuse to launch subsequent AM attempts for that application on
those nodes. If we want to keep it really simple, we could just do literally
that. From there we can sprinkle additional logic to make it a bit more
sophisticated, e.g.: having the blacklisting auto-disable when the percentage
of blacklisted nodes compared to the total active nodes is above some threshold
and/or the app has waited some amount of time for an AM container for the next
attempt.
> 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)
