Hello,
I opened an issue on github in November:
And I'm not ok with the answer I received.
I have 2 namespace in my k8s cluster, I created the CRD in default ns, and the Traefiks as Daemonset in both namespace.
And in both controller, I see it continuously trying to fetch service which either does not belong to his namespace or does not exist anymore in the cluster
Hello @Jean-Philippe,
Thanks for your interest in Traefik!
On the original issue, I see that you have configured the Kubernetes provider in your Traefik service without restricting it to a specific namespace with the appropriate flag (e.g.: --providers.kubernetescrd.namespaces=default,production).
Did you try this option?
It seems to be ok.
But it keeps looking for old services in one of my namespace, like if those deleted resources still exists somewhere in the cluster.
Hello @Jean-Philippe,
It's a bit difficult to help without context on this, can you share your logs?
It would help to see if there are connection issues between Kubernetes and Traefik.