Error when switching Ingress from v1beta1 to v1

[mroloux]

Hello!

• Vote on this issue by adding a 👍 reaction
• To contribute a fix for this issue, leave a comment (and link to your pull request, if you've opened one already)

Issue details

I received an email from Google to point out we're using some deprecated features, which won't be supported by Kubernetes 1.22. So I'm trying to migrate our Ingress controller from networking.k8s.io/v1beta1 to networking.k8s.io/v1.

While applying the new config for the first time, I get an error. Applying it a second time works.

Steps to reproduce

1. create an Ingress controller such as:

export function createIngress(serviceName: string, ip: GlobalAddress, managedCertificate: k8s.apiextensions.CustomResource, nodePort: k8s.core.v1.Service, clusterProvider: k8s.Provider) {
    const ingress = new k8s.networking.v1beta1.Ingress(serviceName + '-ingress', {
            apiVersion: "networking.k8s.io/v1beta1",
            metadata: {
                annotations: {
                    "kubernetes.io/ingress.global-static-ip-name": ip.name,
                    "networking.gke.io/managed-certificates": managedCertificate.metadata.name,
                    "kubernetes.io/ingress.allow-http": "false"
                }
            },
            spec: {
                backend: {
                    serviceName: nodePort.metadata.name,
                    servicePort: 80
                }
            }
        },
        {
            provider: clusterProvider,
            protect: true
        })
}

2. change the Ingress to not use the beta API:

export function createIngress(serviceName: string, ip: GlobalAddress, managedCertificate: k8s.apiextensions.CustomResource, nodePort: k8s.core.v1.Service, clusterProvider: k8s.Provider) {
    new k8s.networking.v1.Ingress(serviceName + '-ingress', {
            metadata: {
                annotations: {
                    "kubernetes.io/ingress.global-static-ip-name": ip.name,
                    "networking.gke.io/managed-certificates": managedCertificate.metadata.name,
                    "kubernetes.io/ingress.allow-http": "false"
                }
            },
            spec: {
                defaultBackend: {
                    service: {
                        name: nodePort.metadata.name,
                        port: {
                            number: 80
                        }
                    }
                }
            }
        },
        {
            provider: clusterProvider,
            protect: true
        })
}

3. run pulumi up

Expected: pulumi up completes successfully
Actual:

error: 1 error occurred:
    	* the Kubernetes API server reported that "default/seatsio-core-v2-staging-na-ingress-e3a28qp7" failed to fully initialize or become live: Ingress.extensions "seatsio-core-v2-staging-na-ingress-e3a28qp7" is invalid: spec: Invalid value: []networking.IngressRule(nil): either `backend` or `rules` must be specified 

Running pulumi up a second time, with the same code, works.

Am I doing something wrong?

[mroloux]

I just found this issue: #1668 - I assume it's related?

[lblackstone]

I just found this issue: #1668 - I assume it's related?

Yeah, that seems possible. Are you blocked by this issue, or are things working aside from that initial error?

[mroloux]

Blocking: yes and no. I only tried upgrading our staging server. Felt wrong to trigger an error on production (who knows what might break).

But sure, technically I could ignore the error while upgrading production.

[lblackstone]

Understood. I've got a potential fix in mind, but will need to do some testing to make sure it doesn't have unintended consequences. I'm hopeful that we can get this fixed in the next couple weeks if you can hold off until then.

[mroloux]

Sounds good 👍