Could we add functionality to the rehoming process so that if rehoming fails the agent falls back to the original on-prem cluster and the UI reports the agents that failed to rehome? Otherwise, there are several manual steps that require a server adm...