[Bug]: Traefik fails after automatic self-restart because of missing CRD about terraform-hcloud-kube-hetzner HOT 4 OPEN

I fixed my current cluster by installing the new CRD and RBAC definitions

# Install Traefik Resource Definitions:
kubectl apply -f https://raw.githubusercontent.com/traefik/traefik/v3.0/docs/content/reference/dynamic-configuration/kubernetes-crd-definition-v1.yml

# Install RBAC for Traefik:
kubectl apply -f https://raw.githubusercontent.com/traefik/traefik/v3.0/docs/content/reference/dynamic-configuration/kubernetes-crd-rbac.yml

and then, in addition, I needed to change all the k8s traefik yaml resource definitions to use apiVersion: traefik.io/v1alpha1.

from terraform-hcloud-kube-hetzner.

Looking around where this autoupdate is defined I found that

• traefik is not autoupdating itself according to a comment found here traefik/traefik#9629 (comment)
• I found traefik_image_tag in kube.tf which says that if not set it will automatically grab the most recent image tag. In addition the deployment seems to have a self-restarting set after a certain amount of time, which together may lead to this self-updating behaviour

from terraform-hcloud-kube-hetzner.

Interesting, maybe a bug should be opened on k3s' side.

from terraform-hcloud-kube-hetzner.

@schlichtanders It could be updated automatically along with k3s. However, if I remember correctly (I need to re-check in the code), we had dissociated traefik from k3s and are now installing it directly via helm. Maybe you were running an old cluster?

from terraform-hcloud-kube-hetzner.