kubernetes-dashboard-kong pod in CrashLoopBackOff after a clean install

[jesuslpm]

I have a clean Docker Desktop Kubernetes where I installed dashboard using:

helm repo add kubernetes-dashboard https://kubernetes.github.io/dashboard/
helm upgrade --install kubernetes-dashboard kubernetes-dashboard/kubernetes-dashboard --create-namespace --namespace kubernetes-dashboard

I then executed kubectl get pod -n kubernetes-dashboard:

NAME                                                  READY   STATUS             RESTARTS      AGE
kubernetes-dashboard-api-768f5bd99f-q4ns5             1/1     Running            0             84s
kubernetes-dashboard-auth-6445f5448f-9ddns            1/1     Running            0             84s
kubernetes-dashboard-kong-75bb76dd5f-qqllc            0/1     CrashLoopBackOff   3 (38s ago)   84s
kubernetes-dashboard-metrics-scraper-8d8778c4-htqz7   1/1     Running            0             84s
kubernetes-dashboard-web-7d77999479-g9k8t             1/1     Running            0             84s

And here is the output of kubectl logs kubernetes-dashboard-kong-75bb76dd5f-qqllc -n kubernetes-dashboard

Defaulted container "proxy" out of: proxy, clear-stale-pid (init)
Error: could not prepare Kong prefix at /kong_prefix: nginx configuration is invalid (exit code 1):
nginx: [warn] the "user" directive makes sense only if the master process runs with super-user privileges, ignored in /kong_prefix/nginx.conf:7
nginx: the configuration file /kong_prefix/nginx.conf syntax is ok
nginx: [emerg] bind() to [::1]:8444 failed (99: Cannot assign requested address)
nginx: configuration file /kong_prefix/nginx.conf test failed

[floreks]

Something occupies the port that kong is trying to use:
nginx: [emerg] bind() to [::1]:8444 failed

[jesuslpm]

How can I debug it myself? as I understand the port is already bound in the container. The Kubernetes cluster is totally empty, just reset, so this should work without my intervention.

[floreks]

With docker desktop it is basically a port forward from the container to the host. Either docker desktop already uses it or something else on your host.

[jesuslpm]

I upgraded to the latest docker desktop and it works now

[karthikGenAI]

I get the same issue in GCP auto-pilot cluster any clues? @floreks @jesuslpm

[floreks]

@karthikGenAI #8765 (comment)