A failover group consists of two or more servers that provide users with a high level of availability in case one server becomes unavailable. When a server in a failover group becomes unavailable, whether for maintenance, a power outage, or any other reason, each Sun Ray Client connected to it reconnects to another server in the failover group that has an existing session for the client's current token. If the client can't find an existing session for the current token, it connects to a server selected by the load balancing algorithm. This server presents a login screen to the user, who then logs in to create a new session. The session on the failed server is lost.

A failover group consists of a primary server and one or more secondary servers, configured with the utreplica command. Each Sun Ray server hosts its own local Sun Ray data store. However, only read access is permitted on the local data stores. Any data changes (write access) are first written on the primary server and later replicated on the secondary servers' Sun Ray data stores.

Servers in a failover group authenticate (or learn to trust) one another by using a common group signature, a key used to sign messages sent between servers in the group. The group signature must be configured, using the utgroupsig command, to be identical on each server.

You can use the utgstatus command or the Servers tab in the Admin GUI to see the status of all servers in a failover group. The following server modes affect how the server participates in a failover group:

When using a failover group, clients are automatically redirected to servers based on load balancing before sessions are created. However, users can use the utselect or utswitch commands for manual redirection.

The utselect GUI is the preferred method to use for server selection. For more information, see the utselect man page.

For more information about how to set up multiple failover groups that use regional hotdesking, see Chapter 6, Hotdesking.