The HA design for the CAFÉ appliance is active-active; since the failover mode is automatic, if one node goes down, the other node will persist to service the client requests without any special configuration.
I would like to succinctly talk about two states followed by the logical sequence of the steps necessary to recover if the active node goes down:
State 1: Failure of one active node
State 2: Failure of all the active nodes
The current state of the CAFÉ nodes can be gathered using the health monitor URL. Open the browser and navigate to https://<CAFE-Loadbalance-IP-or-FQDN>/vcac/services/api/status
, that is https://CAFE.PKCT.LOCAL/vcac/services/api/status
. If both or one of the nodes are active, you should receive a response as shown in the following screenshot: