`minikube image rm` Exits with 0 on Failure

kon-foo commented 2 months ago

What Happened?

Description

When running minikube image rm, the exit code is 0 even though the command fails to remove the image. A failure should result in a non-zero exit code, allowing scripts or CI processes to handle the failure properly.

Steps to reproduce

Load an image and start a pod.

minikube image load alpine:latest
kubectl create deployment alpine-deployment --image=alpine:latest -- sleep 3600

Attempt to remove the image while being in use.
```
minikube image rm alpine:latest
```

Check for the error output: You should see an error similar to:


❗  Failed to remove images for profile minikube error removing images: remove image docker: docker rmi alpine:latest: Process exited with status 1
stdout:

stderr: Error response from daemon: conflict: unable to remove repository reference "alpine:latest" (must force) - container dc9b6c3f872e is using its referenced image 91ef0af61f39

4. Check the exit code of the minikube command:
```bash
echo $?
# Output: 0

Expected Behavior

When minikube image rm encounters an error (such as an image being in use), it should exit with a non-zero status code (e.g. 1), allowing proper error handling in scripts or pipelines.

Actual Behavior

The command exits with a status code of 0, even though it reports an internal failure, which leads to downstream commands executing when they shouldn't.

Suggested Fix

Ensure that minikube image rm propagates the exit status of the underlying image removal process.\

Environment

Minikube Version: v1.33.1
OS: Ubuntu via WSL2 (5.15.153.1-microsoft-standard-WSL2)

Attach the log file

log.txt

Operating System

Ubuntu

Driver

Docker

PyAgni commented 2 months ago

@spowelljr Can I work on this?

spowelljr commented 2 months ago

@PyAgni Yup, comment /assign and it will assign the issue to you

PyAgni commented 2 months ago

/assign

wassafshahzad commented 1 month ago

@PyAgni are you still working on this ?

wassafshahzad commented 1 month ago

@spowelljr I did some RnD on this issue and the problem lies in this block. If the image is not removed it is added into the failed array and the loop continues. It does not error out. I believe this behavior exists to handle multi-node situations. Please do recommend the best way to resolve this. https://github.com/kubernetes/minikube/blob/94d78a95a06c9999601cfcd056d414b89969c85f/pkg/minikube/machine/cache_images.go#L678

kubernetes / minikube