Helm chart 7.0.0-alpha1: Access with port-forward returns 404

[Stolz]

I've installed the dashboard using the latest official helm chart version (7.0.0-alpha1) with these values.yaml file

app:
  security:
    networkPolicy:
      enabled: true
      ingressDenyAll: true
  ingress:
    enabled: false
metrics-server:
  enabled: false
metricsScraper:
  enabled: false

As you can see, metrics and ingress are disabled. I don't have an ingress controller installed in my cluster yet but according to the official documentation I should still be able to access with kubectl port-forward.

The pods and services seem to be successfully created

kubectl get pods -n kubernetes-dashboard
NAME                                                            READY   STATUS    RESTARTS   AGE
kubernetes-dashboard-api-6988dbb97d-7mzqr                       1/1     Running   0           1m
kubernetes-dashboard-cert-manager-76ff4cb656-s7wfw              1/1     Running   0           1m
kubernetes-dashboard-cert-manager-cainjector-55db97cfbb-7hjlh   1/1     Running   0           1m
kubernetes-dashboard-cert-manager-webhook-7b569977cb-2d7jt      1/1     Running   0           1m
kubernetes-dashboard-nginx-controller-65474fdcb8-4n6cp          1/1     Running   0           1m
kubernetes-dashboard-web-58fb55564c-cxkf4                       1/1     Running   0           1m

kubectl get svc -n kubernetes-dashboard
NAME                                              TYPE        CLUSTER-IP        EXTERNAL-IP   PORT(S)          AGE
kubernetes-dashboard-api                          ClusterIP   192.168.50.59     <none>        9000/TCP          2m
kubernetes-dashboard-cert-manager                 ClusterIP   192.168.220.8     <none>        9402/TCP          2m
kubernetes-dashboard-cert-manager-webhook         ClusterIP   192.168.15.227    <none>        443/TCP           2m
kubernetes-dashboard-metrics-scraper              ClusterIP   192.168.152.165   <none>        8000/TCP          2m
kubernetes-dashboard-nginx-controller             ClusterIP   192.168.179.192   <none>        80/TCP,443/TCP    2m
kubernetes-dashboard-nginx-controller-admission   ClusterIP   192.168.28.152    <none>        443/TCP           2m
kubernetes-dashboard-web                          ClusterIP   192.168.238.71    <none>        8000/TCP          2m

But when I try to access with port-forward I get a 404 error

kubectl -n kubernetes-dashboard port-forward svc/kubernetes-dashboard-nginx-controller 8443:443
Forwarding from 127.0.0.1:8443 -> 443
Handling connection for 8443

wget --no-check-certificate https://127.0.0.1:8443/
--2023-11-03 21:25:56--  https://127.0.0.1:8443/
Connecting to 127.0.0.1:8443... connected.
WARNING: cannot verify 127.0.0.1's certificate, issued by 'CN=Kubernetes Ingress Controller Fake Certificate,O=Acme Co':
  Self-signed certificate encountered.
    WARNING: certificate common name 'Kubernetes Ingress Controller Fake Certificate' doesn't match requested host name '127.0.0.1'.
HTTP request sent, awaiting response... 404 Not Found
2023-11-03 21:25:56 ERROR 404: Not Found.

I have tried using https://127.0.0.1:8443, https://127.0.0.1:8443/, https://127.0.0.1:8443/ui, https://127.0.0.1:8443/ui/, https://127.0.0.1:8443/web, https://127.0.0.1:8443/web/, https://127.0.0.1:8443/dashboard, https://127.0.0.1:8443/dashboard/ but they all return 404.

I have also tried with all 8 possible unique combinations of the boolean values listed below, but they all return 404.

    networkPolicy:
      enabled: true
      ingressDenyAll: true
  ingress:
    enabled: true
    hosts:
      - localhost

I have also tried using port-forward svc/kubernetes-dashboard-nginx-controller-admission instead of svc/kubernetes-dashboard-nginx-controller but I still get 404.

Any suggestion of what could be wrong with my approach?

[ahoehma]

I have more or less the same setup ...

cert-manager:
  enabled: false 
nginx:
  enabled: false
metrics-server:
  enabled: false
app:
  ingress:
    enabled: false 

$k get pods,svc,ingress -n kubernetes-dashboard
NAME READY STATUS RESTARTS AGE
pod/kubernetes-dashboard-api-7cc955bcdd-b7kzn 1/1 Running 0 14h
pod/kubernetes-dashboard-metrics-scraper-556c9dd79-n7z6q 1/1 Running 0 14h
pod/kubernetes-dashboard-web-795dc7c5dc-2tsrh 1/1 Running 0 14h

NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
service/kubernetes-dashboard-api ClusterIP XXXXX 9000/TCP 14h
service/kubernetes-dashboard-metrics-scraper ClusterIP XXXXX 8000/TCP 14h
service/kubernetes-dashboard-web ClusterIP XXXXX 8000/TCP 14h

I can only got something working with "kubectl proxy" and then I can open

http://localhost:8001/api/v1/namespaces/kubernetes-dashboard/services/http:kubernetes-dashboard-web:8000/proxy/

But then the page keeps blank and I see a lot of console errors like

404 http://localhost:8001/api/v1/namespaces/kubernetes-dashboard/services/http:kubernetes-dashboard-web:8000/proxy/api/v1/login/status

I believe this is because of the split between web and api services.

No glue how to handle this?! :-(

[comminutus]

i get the same problem. I'm trying to use it with kubectl port-forward and no ingress

[comminutus]

@Stolz can you change this to an issue instead of a Q&A?

[DeadGuruu]

Same problem here