the HealthController needs its own pod in order to incorporate the metrics rework. The centralized declaration, teardown, and setting of metrics will required the MCC, MCD and MCO to "shell" out requests to the MHC (Machine Health Controller) saying things like:

• Set Metric X to value Y
• Declare Metric X for component Y
• Remove Metric X from component Y

This is necessary because if we are looking to centralize the metrics, the issue is that you cannot change the value of a metric whose listener is not in your pod.

Having the MCO, MCD and MCC tell the MHC to update metrics on its next sync will allow for simple uniform register, update, and teardown metric functions all to live in the MHC.