I have a pretty simple nginx ingress setup within the GKE by using the nginx-ingress helm chart:
helm install --namespace nginx-ingress-lb --name nginx-ingress stable/nginx-ingress --set rbac.create=true
All other apps which are also in other namespaces of course use this loadbalancer with the default annotation in each ingress.yaml:
kubernetes.io/ingress.class: "nginx"
Until chart version 1.1.5 and app version 0.21.0 everything was fine. But starting with chart version 1.2.0 and app version 0.22.0 the loadbalancer doesn't forward the traffic anymore. If I do a rollback to 0.21.0 again everything is fine. From my understanding the only thing that had a breaking change was
nginx.ingress.kubernetes.io/rewrite-target
which I don't use.
Any idea would be really appreciated since I don't wanna stick to app version 0.21.0. Thank you in advance!
I did some more tests and it seemed to be an issue with my DNS provider. I let it run for a couple of hours and after a while it worked for some reason which is pretty odd. The only explanation I have is that the name resolution took longer than usual.