[MESOS-6414] cgroups isolator cleanup failed when the hierarchy is cleanup by docker daemon

• Type: Bug
• Status: Resolved
• Priority: Minor
• Resolution: Fixed
• Fix Version/s: 1.2.0
• Component/s: containerization
• Labels: containerizer
• Assignee: Anindya Sinha
• Reporter: Anindya Sinha
• Shepherd: Yan Xu
• Created: 18/Oct/16 22:42
• Updated: 06/Jul/18 21:27
• Resolved: 20/Oct/16 18:29

Description

Now if we launch a docker container in Mesos containerizer, the racing may happen
between docker daemon and Mesos containerizer during cgroups operations.
For example, when the docker container which run in Mesos containerizer OOM exit,
Mesos containerizer would destroy following hierarchies

/sys/fs/cgroup/freezer/mesos/<mesos-cgroup>/<docker-cgroup>
/sys/fs/cgroup/freezer/mesos/<mesos-cgroup>

But the docker daemon may destroy

/sys/fs/cgroup/freezer/mesos/<mesos-cgroup>/<docker-cgroup>

at the same time.

If the docker daemon destroy the hierarchy first, then the Mesos containerizer would
failed during CgroupsIsolatorProcess::cleanup because it could not find that hierarchy
when destroying.

Issue Links

• is related to: MESOS-3086 Allow destroying a cgroup without freezer subsystem.
• is related to: MESOS-5991 Support running docker daemon inside a container using unified containerizer.
• relates to: MESOS-5410 Support cgroup namespace in unified container
• relates to: MESOS-6489 Better support for containers that want to manage their own cgroup.