[
https://issues.apache.org/jira/browse/YARN-1856?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15081359#comment-15081359
]
Karthik Kambatla commented on YARN-1856:
----------------------------------------
Was catching up on YARN-3 (the JIRA that added cgroups) to see why we decided
to not use it for enforcing memory. [~bikassaha] has some [valid
points|https://issues.apache.org/jira/browse/YARN-3?focusedCommentId=13414567&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13414567]
on not letting the kernel (through cgroups) kill processes that go over their
allocated limits.
To get the best of both worlds: I feel we should disable oom_control so the
processes are paused but not killed. Thoughts?
> cgroups based memory monitoring for containers
> ----------------------------------------------
>
> Key: YARN-1856
> URL: https://issues.apache.org/jira/browse/YARN-1856
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: nodemanager
> Affects Versions: 2.3.0
> Reporter: Karthik Kambatla
> Assignee: Varun Vasudev
> Fix For: 2.9.0
>
> Attachments: YARN-1856.001.patch, YARN-1856.002.patch,
> YARN-1856.003.patch, YARN-1856.004.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
