change HA carp state checks
We've had many customers in the last few months that are finding both nodes in carp MASTER state. This ticket is to request that the nodes check their carp states and see if they "agree" (that is, one is MASTER and one BACKUP). If not, throw a critical alert in the WebGUI, make the HA icon be red (HA will not work if both are already in MASTER state), and send an email.
Because of this condition, I'm finding customers that have no alerts or any indication that HA won't work (no alerts, HA icon is green, etc.) and then the failover fails.
This causes a production down situation for HA if a customer experiences a failure of the active node and the passive node's carp state is MASTER due to this condition in the external network.