Hi Jennifer,
The manual failover refers to the management/configuration portion of the ACS deployment. In a deployment, changes and management is done only through the primary node which in return pushes the changes to the rest of the nodes in the deployment. So, if the primary node fails then you have to manually promote a secondary node if you want to perform any changes.
With regards to AAA functionality (TACACS+ and/or RADIUS), the redundancy is there and each node from the deployment will provide that functionality without the need to manually failover. What you need to make sure though is that your NAD (switch, router, firewall, etc) has each ACS node listed as a RADIUS/TACACS+ server. That way if server #1 becomes unavailable the NAD will move down to the 2nd one configured.
I hope this helps!
Thank you for rating helpful posts!