You are viewing a plain text version of this content. The canonical link for it is here.
Posted to yarn-issues@hadoop.apache.org by "Miklos Szegedi (JIRA)" <ji...@apache.org> on 2017/03/29 19:34:41 UTC
[jira] [Updated] (YARN-5301) NM mount cpu cgroups failed on some
system
[ https://issues.apache.org/jira/browse/YARN-5301?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Miklos Szegedi updated YARN-5301:
---------------------------------
Attachment: YARN-5301.000.patch
The attached patch fixes the issue by identifying any premounted cgroups and the subsystem set for each mount point. If the user specifies the mount option that means that the existing mount point might not be usable (access, design, etc.), so we remount it with the existing subsystem set (cpu, cpuacct in most cases). I tested it on CentOS6, Ubuntu16 with kernel 4.4. I am testing it on RHEL7 now.
I also updated the unit tests and fixed a few compiler warnings in those files.
> NM mount cpu cgroups failed on some system
> ------------------------------------------
>
> Key: YARN-5301
> URL: https://issues.apache.org/jira/browse/YARN-5301
> Project: Hadoop YARN
> Issue Type: Bug
> Reporter: sandflee
> Assignee: Miklos Szegedi
> Attachments: YARN-5301.000.patch
>
>
> on ubuntu with linux kernel 3.19, , NM start failed if enable auto mount cgroup. try command:
> ./bin/container-executor --mount-cgroups yarn-hadoop cpu=/cgroup/cpu fail
> ./bin/container-executor --mount-cgroups yarn-hadoop cpu,cpuacct=/cgroup/cpu succ
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org