With Control-M Server  setup in High Availability mode, the secondary Control-M Server should not be discovered in the CCM. As the Primary platform is discovered, when a failover occurs, the secondary will come online. 

When manually shutting down the Control-M Server components on the primary, this does not trigger an automatic failover as it recognizes a "clean" manual intervention. If the primary platform processes dies or become unresponsive, or gets disconnected from the network, then there will likely be failure in a life check and a database check from the secondary Configuration Agent, which will then trigger the automatic failover. 

In order to force an Automatic failover in High Availability, manually disrupt the communication between the primary and secondary Configuration Agent processes and from the database. One of the following actions could trigger the automatic failover:

1. Reboot the primary host.
2. Disable or stop the network interface on the primary host and network interface to the database.
4. Issue "kill -9" to the Control-M Server Configuration Agent process on the primary host, followed by the remaining Control-M Server processes. If the Control-M Server is on Windows platform, ensure to change the Control-M Server's services recovery settings to "Take No Action" first(please make sure to set back to default settings later), so that these services will not be restarted by the Windows OS automatically, when the service processes are killed manually.