Is there an existing issue for this?
Current Behavior
The first time a request using a reflector is received, we initialize the reflector for the given resource/selectors.
If the fetching of the initial resource list fails, the reflector does not return as ready, therefore the client hangs indefinitely. As a result, the policy waits till the Kubernetes API server timeout is reached.
Expected Behavior
If the reflector initialization fails, the client should return an error.