[
https://issues.apache.org/jira/browse/MESOS-2154?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15051749#comment-15051749
]
Steve Niemitz commented on MESOS-2154:
--------------------------------------
The issue with only setting the CFS quota on the command line is that the quota
can change over time as tasks add/leave the container. Correctly supporting
CFS requires directly updating the cgroup to handle this case, the same way
cpu.shares is handled.
We've been running a version of mesos w/ a custom CFS patch [1] applied for ~6
months now, I had a review open a long time ago that I gave up on trying to get
upstream, but I think I'm going to revisit it now.
[1] https://reviews.apache.org/r/33174/diff/2/
> Port CFS quota support to Docker Containerizer
> ----------------------------------------------
>
> Key: MESOS-2154
> URL: https://issues.apache.org/jira/browse/MESOS-2154
> Project: Mesos
> Issue Type: Improvement
> Components: docker, isolation
> Affects Versions: 0.21.0
> Environment: Linux (Ubuntu 14.04.1)
> Reporter: Andrew Ortman
> Assignee: haosdent
> Priority: Minor
>
> Port the CFS quota support the Mesos Containerizer has to the Docker
> Containerizer. Whenever the --cgroup_enable_cfs flag is set, the Docker
> Containerizer should update the cfs_period_us and cfs_quota_us values to
> allow hard CPU capping on the container.
> Current workaround is to pass those values as LXC configuration parameters
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
