[
https://issues.apache.org/jira/browse/YARN-8648?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16584600#comment-16584600
]
Eric Yang commented on YARN-8648:
---------------------------------
I am in favor of minimal fix at this time. Let docker be docker seems to have
possibility to introduce risk where container-executor usage is either
unaccounted, or Kernel underflow occurs. When cgroup is managed separately,
there is a chance that YARN and Docker don't agree on actual usage. (i.e. 2GB
allocated to container, but container-executor used 5mb, and docker used 2GB.)
Miscalculated number can cause Kernel underflow. If YARN uses a flat namespace
like docker does, then it will be safe to let docker be docker. In today's
reality, there are people out there that don't have cgroup v2 in their kernel.
This means it is less likely to be a feasible option to do major change to yarn
cgroup, although it would be a great option to have two years down the road to
support cgroup v2.
> Container cgroups are leaked when using docker
> ----------------------------------------------
>
> Key: YARN-8648
> URL: https://issues.apache.org/jira/browse/YARN-8648
> Project: Hadoop YARN
> Issue Type: Bug
> Reporter: Jim Brennan
> Assignee: Jim Brennan
> Priority: Major
> Labels: Docker
>
> When you run with docker and enable cgroups for cpu, docker creates cgroups
> for all resources on the system, not just for cpu. For instance, if the
> {{yarn.nodemanager.linux-container-executor.cgroups.hierarchy=/hadoop-yarn}},
> the nodemanager will create a cgroup for each container under
> {{/sys/fs/cgroup/cpu/hadoop-yarn}}. In the docker case, we pass this path
> via the {{--cgroup-parent}} command line argument. Docker then creates a
> cgroup for the docker container under that, for instance:
> {{/sys/fs/cgroup/cpu/hadoop-yarn/container_id/docker_container_id}}.
> When the container exits, docker cleans up the {{docker_container_id}}
> cgroup, and the nodemanager cleans up the {{container_id}} cgroup, All is
> good under {{/sys/fs/cgroup/hadoop-yarn}}.
> The problem is that docker also creates that same hierarchy under every
> resource under {{/sys/fs/cgroup}}. On the rhel7 system I am using, these
> are: blkio, cpuset, devices, freezer, hugetlb, memory, net_cls, net_prio,
> perf_event, and systemd. So for instance, docker creates
> {{/sys/fs/cgroup/cpuset/hadoop-yarn/container_id/docker_container_id}}, but
> it only cleans up the leaf cgroup {{docker_container_id}}. Nobody cleans up
> the {{container_id}} cgroups for these other resources. On one of our busy
> clusters, we found > 100,000 of these leaked cgroups.
> I found this in our 2.8-based version of hadoop, but I have been able to
> repro with current hadoop.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
