node-local-dns ConfigMap deletion

[jem41]

Hello,

We use Azure Kubernetes Service with Azure CNI Networking. With the addonmanager.kubernetes.io/mode: Reconcile label on the node-local-dns' ConfigMap, the resource gets deleted shortly after a Helm upgrade. Is it intentional for the ConfigMap to be removed? If so, how are these values read if nodes scale out or a pod restarts?

We recently added ArgoCD to our cluster and the ConfigMap's deletion causes node-local-dns to perpetually be OutOfSync. If deletion is intentional, would you accept a PR to allow specifying annotations to the ConfigMap? That would allow us to ignore the resource in ArgoCD for sync purposes.