quick question...
have you checked to see if the controller has failed over...
if you logon to the running controller you will be able to see its cluster failover status cf status will work, but the CLI itself should show you are in failover mode as soon as you logon.
I assume that this HA configuration has worked previously?