[
https://issues.apache.org/jira/browse/YARN-5827?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15683535#comment-15683535
]
Shane Kumpf commented on YARN-5827:
-----------------------------------
Can you give me the steps you took to reconfigure docker? What does {{ps -ef |
grep docker}} return? Do you see the exec-opt in the ps output?
_systemd cgroup_ in the error output indicates that systemd is in use still, so
it doesn't sound like docker was reconfigured correctly. Let me know the steps
you took and I can try to advise.
> DockerLinuxContainer can't limit the cpu usage
> ----------------------------------------------
>
> Key: YARN-5827
> URL: https://issues.apache.org/jira/browse/YARN-5827
> Project: Hadoop YARN
> Issue Type: Bug
> Components: nodemanager
> Affects Versions: 2.8.0
> Environment: CentOS7.1
> Reporter: zhengchenyu
> Original Estimate: 240h
> Remaining Estimate: 240h
>
> When I use DockerLinuxContainer, I found that the function of limiting cpu
> usage doesn't make effect. Because DockerLinuxContainer regard the exitCode
> of "docker inpsect" as pid. But When I got this pid, the command of
> "launch.sh" has started, then we can't add the child process or threads into
> the cgroup, so the restrictions of cpu usage doesn't make effect。
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
