A failover for a DB cluster promotes one of the Aurora Replicas (read-only instances) in the DB cluster to be the primary instance (the cluster writer).
Amazon Aurora will automatically fail over to an Aurora Replica, if one exists, when the primary instance fails. You can force a failover when you want to simulate a failure of a primary instance for testing. Because each instance in a DB cluster has its own endpoint address, you will need to clean up and re-establish any existing connections that use those endpoint addresses when the failover is complete.
For more information on Amazon Aurora, see Aurora on Amazon RDS in the Amazon RDS User Guide.