[
https://issues.apache.org/jira/browse/YARN-2889?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16270401#comment-16270401
]
Konstantinos Karanasos commented on YARN-2889:
----------------------------------------------
To add to what [~asuresh] described, an extra reason for such a limit would be
malicious AMs.
When an AM requests opportunistic containers, we are purposely not checking
whether there are resources to run such containers (they will be queued in the
NMs if there are no resources). If an AM (either due to bad design or being
malicious) requests too many opportunistic containers, these containers will
flood the system and will possibly block O containers of other apps from being
executed.
What we did as a first step with YARN-7258 (as Arun described) and what this
JIRA would do is to not allow such a situation by imposing limits on the number
of O containers that a single AM can ask.
So, in your case, if your AMs ask for a reasonable amount of opportunistic
containers (and below the configured limit), there should not be a problem.
> Limit in the number of opportunistic container requests per AM
> --------------------------------------------------------------
>
> Key: YARN-2889
> URL: https://issues.apache.org/jira/browse/YARN-2889
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: nodemanager, resourcemanager
> Reporter: Konstantinos Karanasos
> Assignee: Arun Suresh
>
> We introduce a way to limit the number of queueable requests that each AM can
> submit to the LocalRM.
> This way we can restrict the number of queueable containers handed out by the
> system, as well as throttle down misbehaving AMs (asking for too many
> queueable containers).
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
