If more than one server has been defined for any type of authentication, you can define the priority of the servers in the case of failover.
In the event of a failover of the main RADIUS server—if there is no response after the set number of retries—then the other servers in the list will be polled on a round-robin basis until a server responds.
If all defined RADIUS servers fail to respond, a critical message is generated in the logs.
To Define the RADIUS Server Priority for RADIUS Redundancy: