event for test-webserver-e9de1b49-9d28-4ca5-9a03-5f244abbe4f2: {kubelet tmp-node-e2e-edd441f7-cos-beta-121-18867-0-24} FailedCreatePodSandBox: (combined from similar events): Failed to create pod sandbox: rpc error: code = Unknown desc = failed to create containerd task: failed to start shim: start failed: dial unix �{"version":3,"address":"unix:///run/containerd/s/148ce26d953fb243c85ce370a41dca14abc101ac3dc2423e5fc3544479dced42","protocol":"ttrpc"}: connect: invalid argument: unknown
Failed to create pod sandbox: rpc error: code = Unimplemented desc = failed to create containerd task: failed to start shim: start failed: unsupported shim version (3): not implemented
Which jobs are failing?
Which tests are failing?
All the tests are failing with the kubetest timeout
Since when has it been failing?
Since k/k commit e1fc73d
k/test-infra commit
Testgrid link
https://testgrid.k8s.io/containerd-presubmits#pull-containerd-node-e2e-1-7
Reason for failure (if possible)
which is a shim version mismatch issue
It seems that containerd on the host OS (cos) is being used here. Because in the job artifacts there is no
containerd-installation.logwhich is created as part of installing the specified version of containerd.Failed bucket: https://gcsweb.k8s.io/gcs/kubernetes-ci-logs/pr-logs/pull/containerd_containerd/11437/pull-containerd-node-e2e-1-7/1894567245917458432/
Passing bucket: https://gcsweb.k8s.io/gcs/kubernetes-ci-logs/pr-logs/pull/containerd_containerd/11422/pull-containerd-node-e2e-1-7/1894077257589198848/artifacts/tmp-node-e2e-9a6c73d3-cos-beta-117-18613-164-38/
Anything else we need to know?
No response
Relevant SIG(s)
/sig node