Skip to content

Using systemd cgroup driver, failed to get stats for kubepods.slice/ #59993

@derekwaynecarr

Description

@derekwaynecarr

Is this a BUG REPORT or FEATURE REQUEST?:
Follow-on issue from: #57802

See the following when using cgroup-driver=systemd every 10s

$ cat /tmp/kubelet.log | grep Failed
E0216 18:02:29.460244     302 summary.go:102] Failed to get system container stats for "kubepods.slice/": failed to get cgroup stats for "kubepods.slice/": failed to get container info for "kubepods.slice/": unknown container "kubepods.slice/"

Issue is cAdvisor expects a leading / and never a trailing slash, but the expansion method used to do the conversion is omitting leading slash, and always having a trailing slash.

https://github.com/opencontainers/runc/blob/master/libcontainer/cgroups/systemd/apply_systemd.go#L399

we can probably look at a change in runc, or just fix-up the adapt function in kubelet in interim or fixup here:

https://github.com/kubernetes/kubernetes/blob/master/pkg/kubelet/cm/cgroup_manager_linux.go#L60

/kind bug
/sig node
/priority important-critical-soon

Metadata

Metadata

Labels

kind/bugCategorizes issue or PR as related to a bug.priority/important-soonMust be staffed and worked on either currently, or very soon, ideally in time for the next release.sig/nodeCategorizes an issue or PR as relevant to SIG Node.

Type

No type

Projects

No projects

Milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions