Argo-cd: ArgoCD fails to create the nginx-ingress webhook

[estahn]

Describe the bug

ArgoCD fails to create the nginx-ingress webhook due to the cluster role not being installed.

See slack conversation here.

https://github.com/helm/charts/blob/master/stable/nginx-ingress/templates/admission-webhooks/job-patch/clusterrole.yaml

To Reproduce

Install nginx-ingress with webhook enabled.

Expected behavior

Install webhook.

Version

v1.5.1+8a3b36b

Logs

The output on the pods seems to be fine, so I guess the hooks are not executed correctly:

argocd@argocd-repo-server-5df4d5fb89-dwpqr:/tmp/https:__kubernetes-charts.storage.googleapis.com_$ helm template nginx-ingress-1.39.0.tgz --set controller.admissionWebhooks.enabled=true | grep -A3 job-patch/clusterrole.yaml
# Source: nginx-ingress/templates/admission-webhooks/job-patch/clusterrole.yaml
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
argocd@argocd-repo-server-5df4d5fb89-dwpqr:/tmp/https:__kubernetes-charts.storage.googleapis.com_$ helm2 template nginx-ingress-1.39.0.tgz --set controller.admissionWebhooks.enabled=true | grep -A3 job-patch/clusterrole.yaml
# Source: nginx-ingress/templates/admission-webhooks/job-patch/clusterrole.yaml
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
argocd@argocd-repo-server-5df4d5fb89-dwpqr:/tmp/https:__kubernetes-charts.storage.googleapis.com_$

[jdfalk]

Argocd doesn't follow the spirit of how helm deploys things. It creates the service account and instantly deletes it. There is another bug open on this because it also happens for the prometheus operator. I really wish they would make some progress as this is a huge hassle to run these things manually then have argo take over.

See: https://github.com/argoproj/argo-cd/issues/3502

[alexmt]

I think this is a duplicate of https://github.com/argoproj/argo-cd/issues/2737 . I really hope to get it fixed in 1.7 release.

Closing this ticket as a duplicate. @estahn let me know please if I'm missing something and ticket has to be reopened

[foxracle]

@alexmt Is there a timeline for this fix to be released? I am still running into this issue with ingress Nginx.

argocd version: 1.7.7