Documentation

Balancer: Failover

This balancer type is intended to provided a failover or backup server to fulfill requests only during failure of the machine(s) in the backend pool.

This balancer is very useful when the tendency is to use single backends to accomodate an existing code base using tracking sessions, log-ins, etc. In such scenarios, just-in-case balancing might be required for situations where the main application server fails.

If multiple entries are defined, the requests will be serverd through the first one on the list. When a failure is detected, Cherokee will kick requests to the next entry on the list for processing. Upon the application server returning, Cherokee will resume sending traffic to the original main backend application server for processing.

The only thing needed to configure this balancer is a list of information sources. At least one must be selected in order for this to work.