Crashloopbackoff after update to 8.5.1

[Teonyr]

Describe your Issue

After updating the chart from 8.5.0 to 8.5.1 the nextcloud pod won't successfully start.
It's entering a Crashloopbackoff-state.

Logs and Errors

The pods log itself states "Upgrading nextcloud from 32.0.0.13 ...."

kubectl describe pod nextcloud-6845ccd787-mrlbz returns this:

Normal   Scheduled  8m44s                  default-scheduler  Successfully assigned default/nextcloud-6845ccd787-mrlbz to bastet
Normal   Pulled     8m44s                  kubelet            Container image "quay.io/k8tz/k8tz:0.18.0" already present on machine
Normal   Created    8m44s                  kubelet            Created container: k8tz
Normal   Started    8m44s                  kubelet            Started container k8tz
Normal   Pulling    8m43s                  kubelet            Pulling image "nextcloud:32.0.1-apache"
Normal   Pulled     7m20s                  kubelet            Successfully pulled image "nextcloud:32.0.1-apache" in 1m22.692s (1m22.692s including waiting). Image size: 523993624 bytes.
Warning  Unhealthy  4m34s (x9 over 7m4s)   kubelet            Liveness probe failed: Get "http://10.42.1.195:80/status.php": dial tcp 10.42.1.195:80: connect: connection refused
Normal   Started    3m4s (x5 over 7m20s)   kubelet            Started container nextcloud
Normal   Killing    2m34s (x5 over 6m44s)  kubelet            Container nextcloud failed liveness probe, will be restarted
Normal   Created    2m4s (x6 over 7m20s)   kubelet            Created container: nextcloud
Warning  Unhealthy  2m4s (x32 over 7m8s)   kubelet            Readiness probe failed: Get "http://10.42.1.195:80/status.php": dial tcp 10.42.1.195:80: connect: connection refused
Normal   Pulled     2m4s (x5 over 6m13s)   kubelet            Container image "nextcloud:32.0.1-apache" already present on machine

I guess the upgrading isn't fast enough, so the liveness probe fails and therefore the pod gets restarted, which results in a loop. (hence Crashloopbackoff)

Describe your Environment

• Kubernetes distribution: k3s

• Helm Version (or App that manages helm): version.BuildInfo{Version:"v3.19.0", GitCommit:"3d8990f0836691f0229297773f3524598f46bda6", GitTreeState:"clean", GoVersion:"go1.24.6"}

• Helm Chart Version: 8.5.1

• values.yaml:

# paste your values.yaml (anonymize any sensitive data)

Additional context, if any

Not posting my values.yaml or other infos, since I think they're unnecessary.
Reverting back to 8.5.0 makes the nextcloud pod start up again.

Thx for your help & work.
If you need more info, please ask and I'll try provide it.

[mreiche]

Check the output of http://10.42.1.195:80/status.php, this may tell you the actual issue.

[Teonyr]

Since the issue is almost two months old, I just solved it in the meantime by temporarily disabling the probes for the update and enabling them again afterwards.

But thanks for your reply.
Maybe it'll help others.