Can't access public IP of AKS ingress

Question

im struggling to get my AKS cluster to work. I can't access my public IP for my service (using nginx-ingress-controller ) I triess the azure load balancer ingress but that wasnt working either I created a static public IP in my azure console and assigned it to my nginx ingress conrtoller - that IP doesn't work either I tried also deleting the kube-proxy pods, as some people complained that these wouldn't update after recreating their ingress, so they had to recreated by deleting them. It's not an issue of no ingress rules being setup, because I'm getting a connection Time out (IP can't even be reached) rather than a 404 error

Here's my ingress config:

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: app
  annotations:
    appgw.ingress.kubernetes.io/ssl-redirect: "true"
spec:
  tls:
    - hosts:
        - api.app
      secretName: tls
  ingressClassName: nginx
  rules:
    - http:
        paths:
          - path: /
            pathType: Prefix
            backend:
              service:
                name: app
                port:
                  number: 8080

Here's my services

$ kubectl get svc -A

default       kubernetes                           ClusterIP      10.0.0.1       <none>         443/TCP                         2d1h
default       app                                  LoadBalancer   10.0.85.118    20.81.3.210    8080:30578/TCP,6565:31233/TCP   171m
ingress       ingress-nginx-controller             LoadBalancer   10.0.21.143    20.121.107.7   80:31325/TCP,443:31584/TCP      18m

I can reach my app at the public IP for the service (load balancer config)

http://20.81.3.210:8080 - 401 API error, as expected

but not at the ingress https://20.121.107.7 - connection timeout

---

I heard sometimes the kube proxies fail to update after setting up an ingress, so I tried deleting the kube-proxy pods (and letting them re-create) but that didn't fix anything.

Also tried assigning the ingress to a manually created public-IP address (assigned by the Azure platform) but that didn't help either

Answer 1

Tobi,

Check out this github issue: https://github.com/Azure/AKS/issues/2903

If you are running latest clstuer 1.24, you might want to downgrade to 1.23 for now because yes, there is an issue with the health probes and Azure cluster.