Got confusing errors when installing kubelet component

[astraw99]

What happened?

When installing the latest kubelet component, got the error:

Sep 15 13:20:23 worker-1 kubelet[8014]: I0915 13:20:23.019025    8014 state_mem.go:36] "Initialized new in-memory state store"
Sep 15 13:20:23 worker-1 kubelet[8014]: I0915 13:20:23.019075    8014 util_unix.go:104] "Using this endpoint is deprecated, please consider using full URL format" endpoint="" URL="unix://"
Sep 15 13:20:23 worker-1 kubelet[8014]: I0915 13:20:23.019875    8014 dynamic_cafile_content.go:157] "Starting controller" name="client-ca-bundle::/var/lib/kubernetes/ca.crt"
Sep 15 13:20:23 worker-1 kubelet[8014]: W0915 13:20:23.020243    8014 logging.go:59] [core] [Channel #1 SubChannel #2] grpc: addrConn.createTransport failed to connect to {
Sep 15 13:20:23 worker-1 kubelet[8014]:   "Addr": "",
Sep 15 13:20:23 worker-1 kubelet[8014]:   "ServerName": "",
Sep 15 13:20:23 worker-1 kubelet[8014]:   "Attributes": null,
Sep 15 13:20:23 worker-1 kubelet[8014]:   "BalancerAttributes": null,
Sep 15 13:20:23 worker-1 kubelet[8014]:   "Type": 0,
Sep 15 13:20:23 worker-1 kubelet[8014]:   "Metadata": null
Sep 15 13:20:23 worker-1 kubelet[8014]: }. Err: connection error: desc = "transport: Error while dialing dial unix: missing address"
Sep 15 13:20:23 worker-1 kubelet[8014]: E0915 13:20:23.020961    8014 run.go:74] "command failed" err="failed to run Kubelet: unable to determine runtime API version: rpc error: code = Unavailable desc = connection error: desc = \"transport: Error while dialing dial unix: missing address\""
Sep 15 13:20:23 worker-1 systemd[1]: kubelet.service: Main process exited, code=exited, status=1/FAILURE
Sep 15 13:20:23 worker-1 systemd[1]: kubelet.service: Failed with result 'exit-code'.

Then added the flag --container-runtime-endpoint=unix:///var/run/docker.sock, got error:

Sep 15 14:55:42 worker-1 kubelet[16034]: I0915 14:55:42.281052   16034 state_mem.go:36] "Initialized new in-memory state store"
Sep 15 14:55:42 worker-1 kubelet[16034]: W0915 14:55:42.281598   16034 logging.go:59] [core] [Channel #1 SubChannel #2] grpc: addrConn.createTransport failed to connect to {
Sep 15 14:55:42 worker-1 kubelet[16034]:   "Addr": "/var/run/docker.sock",
Sep 15 14:55:42 worker-1 kubelet[16034]:   "ServerName": "/var/run/docker.sock",
Sep 15 14:55:42 worker-1 kubelet[16034]:   "Attributes": null,
Sep 15 14:55:42 worker-1 kubelet[16034]:   "BalancerAttributes": null,
Sep 15 14:55:42 worker-1 kubelet[16034]:   "Type": 0,
Sep 15 14:55:42 worker-1 kubelet[16034]:   "Metadata": null
Sep 15 14:55:42 worker-1 kubelet[16034]: }. Err: write unix @->/var/run/docker.sock: write: broken pipe
Sep 15 14:55:42 worker-1 kubelet[16034]: E0915 14:55:42.281662   16034 run.go:74] "command failed" err="failed to run Kubelet: unable to determine runtime API version: rpc error: code = Unavailable desc = write unix @->/var/run/docker.sock: write: broken pipe"
Sep 15 14:55:42 worker-1 systemd[1]: kubelet.service: Main process exited, code=exited, status=1/FAILURE
Sep 15 14:55:42 worker-1 systemd[1]: kubelet.service: Failed with result 'exit-code'.

It is confusing to see these error msgs at the first glance, need to make it more clearly.

What did you expect to happen?

Add the required field validation, and add more supportive error message.

How can we reproduce it (as minimally and precisely as possible)?

Download the latest v1.25.0 kubelet binary, and install it with systemd service or just manually start the binary,
will reproduce these errors.

Kubernetes version

$ kubectl version
v1.25.0

OS version

$ uname -a
Linux worker-1 4.15.0-192-generic #203-Ubuntu SMP Wed Aug 10 17:40:03 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux

[k8s-ci-robot]

@astraw99: This issue is currently awaiting triage.

If a SIG or subproject determines this is a relevant issue, they will accept it by applying the triage/accepted label and provide further guidance.

The triage/accepted label can be added by org members by writing /triage accepted in a comment.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[astraw99]

/sig node

[SergeyKanzhelev]

kubelet 1.24+ is not working with the docker, unless using https://github.com/Mirantis/cri-dockerd Are you using it? If not, try switching to other available container runtimes. Please re-open if you have further questions

/close

[k8s-ci-robot]

@SergeyKanzhelev: Closing this issue.

In response to this:

kubelet 1.24+ is not working with the docker, unless using https://github.com/Mirantis/cri-dockerd Are you using it? If not, try switching to other available container runtimes. Please re-open if you have further questions

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[astraw99]

@SergeyKanzhelev Yes, I got to know the cri-dockerd, so need to add runtime validation by the fix PR: #112542
PTAL thanks.