Ingress-nginx: GCE error unable to create HTTPS LB

[tonglil]

I am getting a strange error that I cannot debug when I tried to create an ingress today (that was working a couple days ago):

googleapi: Error 400: Invalid value for field 'namedPorts[2].port': '0'. Must be greater than or equal to 1, invalid

This Stackoverflow answer doesn't work for me as there are no port mappings on my instance groups.

Any insight into this?

[tonglil]

This is the manifest:

---
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
 name: test
spec:
 backend:
   serviceName: echo-app
   servicePort: 88
---
apiVersion: v1
kind: Service
metadata:
  name: echo-app
spec:
  type: NodePort
  selector:
    app: echo-app
  ports:
    - name: http
      port: 88
      protocol: TCP
      targetPort: 8000

kubectl version:

Client Version: version.Info{Major:"1", Minor:"7", GitVersion:"v1.7.5", GitCommit:"17d7182a7ccbb167074be7a87f0a68bd00d58d97", GitTreeState:"clean", BuildDate:"2017-08-31T09:14:02Z", GoVersion:"go1.8.3", Compiler:"gc", Platform:"darwin/amd64"}
Server Version: version.Info{Major:"1", Minor:"7", GitVersion:"v1.7.5", GitCommit:"17d7182a7ccbb167074be7a87f0a68bd00d58d97", GitTreeState:"clean", BuildDate:"2017-08-31T08:56:23Z", GoVersion:"go1.8.3", Compiler:"gc", Platform:"linux/amd64"}

[nicksardo]

Hey @tonglil
Error events can appear for any/all ingresses regardless of the ingress that caused it. I'm guessing a service without type NodePort was referenced in one of your ingress manifests.

[tonglil]

@nicksardo the strange thing was that the manifests didn't change (for the 2 ingress in the cluster).

It's likely it was a temporary issue. I can reopen if I encounter it again.

Thanks for following up!

[lvnilesh]

It was working for me two hours ago. I changed nothing and now I see. Hope it is temporary.

googleapi: Error 400: Invalid value for field 'namedPorts[0].port': '0'. Must be greater than or equal to 1, invalid

[dfang]

same here

[aledbf]

@dfang please open a new issue in https://github.com/kubernetes/ingress-gce

[nicksardo]

This comes up all the time. It always happens when one of your ingress objects references a service which doesn't have type NodePort. https://github.com/kubernetes/ingress-gce/issues/75

[ptagr]

got the error. Service had to be NodePort

[Freyert]

I ran into this issue today, and the solution was to delete and re-install the NGINX ingress controller. I can route to ClusterIP services again!

I suppose this is indicative of something sinister though.

This is using stable/nginx-ingress-0.28.2 from https://github.com/helm/charts