Data Guard Log Management Tips Oracle Database Tips by Donald BurlesonDecember 9, 2015

Oracle Data GuardChapter 6 - Switchover and Failover

Role Transition Using Failover

After the failover, the original primary database is removed from the Data Guard configuration. Other standby databases may or may not be affected depending on the type of failover. Based on the amount of data loss and the impact on other standby databases in the Data Guard configuration, the failover operation can be classified into two categories.

Graceful Failover

A graceful failover is similar to a switchover operation except that in a graceful failover the original primary database is discarded. Graceful failover reduces the amount of data loss and the impact on other standby databases. A graceful failover is ideally suited for physical standby databases, in which standby redo logs can be created and maintained. Graceful failover supports the recovery of all possible redo data from the standby redo log file or the archived redo log file before switching the standby database over to the primary database.

Depending on the type of protection mode, failover can occur with no data loss or minimal data loss. In some cases, even if the primary database is running in maximum protection mode, it may not be possible to recover all the redo data from the standby redo log file. One such situation is a corrupt standby redo log file during the disaster of the primary database.  Fortunately, the probability of these two events occurring simultaneously is negligible. Details of graceful failover will be presented later in this chapter.

Forced Failover

A forced failover operation can result in data loss. However, the amount of data loss can be minimized by copying the online log files and the archived redo log files from the original primary database and registering with the standby database. As a result, the amount of data that can be saved depends on the kind of failure on the primary database.