The original ticket asked to support a "management cluster". Our docs may want to use this terminology, and an external Kiali is just one aspect. We may want to document a best practice involving offloading of all the observability components, Prometheus, Tracing, Perses etc.