Is the controller chasing its tail here? Updating status. which causes a conflict, etc.?
I don't know what's causing this. We do update the objects status during the reconcile loop - however the update is supposed to update the instance we are using so I'm not sure why we'd encounter a conflict. I wonder if there's caching in the client code that's causing this. Maybe we should update kubebuilder and see if this resolves it. I did poke through operators and found that at least knative's update method handles conflicts magically.
I don't know what's causing this. We do update the objects status during the reconcile loop - however the update is supposed to update the instance we are using so I'm not sure why we'd encounter a conflict. I wonder if there's caching in the client code that's causing this. Maybe we should update kubebuilder and see if this resolves it. I did poke through operators and found that at least knative's update method handles conflicts magically.