In Data Guard, we can simply distinguish switchover and failover as planned and unplanned role transitions. A switchover is a planned role transition between the primary database and one of its standby databases. Switchover can be considered to reduce downtime during scheduled maintenance on the primary system or to test stability for future role transitions. Switchover guarantees no data loss. Using switchover, the primary database can transit to a standby role, and the standby database can transit to the primary role at any point of time. Switchover can be performed through Cloud Control, the Data Guard broker command-line interface, or by issuing SQL*Plus commands.

Once the standby database is configured and is functioning properly, you can test switchover. Switchover is used to reduce primary database downtime during any OS or hardware upgrades, which require an extended outage. A switchover allows the primary database to switch roles with its standby database. Once the maintenance on the primary server has been performed, you can switch the databases back to their original roles.

In the case of primary database failure, you need to perform failover to transit from the standby database role to the primary role. After a failover, the original primary database cannot participate in the Data Guard configuration without the use of flashback. So if the original primary database is still accessible, you should always consider a switchover first.

A failover is performed when the production database (all instances of an Oracle RAC production database) fails. By performing failover, one of the standby databases is transitioned to take over the production role, allowing business operations to continue. Once the failover is complete and applications have resumed, the administrative staff can turn its attention to resolving the problems with the failed system. Failover may or may not result in data loss depending on the Data Guard protection mode in effect at the time of the failover.