Oracle7 Server Concepts

Survivability

In the event of a power failure, hardware failure, or any other system-interrupting disaster, Oracle7 release 7.3 offers the standby database feature. The standby database is intended for sites where survivability and disaster recovery are of paramount importance.

For information about creating and maintaining a standby database, see the Oracle7 Server Administrator's Guide.

Planning for Disaster Recovery

The only way to ensure rapid recovery from a system failure or other disaster is to plan carefully. You must have a set plan with detailed procedures. Whether you are implementing a standby database or you have a single system, you must have a plan for what to do in the event of a catastrophic failure.

See Chapter 24, "Database Recovery", for information about recovering a database.

Standby Database

Release 7.3 provides a reliable and supported mechanism for implementing a standby database system to facilitate quick disaster recovery. The scheme uses a secondary system on duplicate hardware, maintained in a constant state of media recovery through the application of log files archived at the primary site. In the event of a primary system failure, the standby can be activated with minimal recovery, providing immediate system availability. There are new commands and internal verifications for operations involved in the creation and maintenance of the standby system, improving the reliability of the disaster recovery scheme.

A standby database involves two databases: a primary database and a standby database. The primary database is the production database that is in use. The standby database is a copy of the production database, ideally located on a separate machine. The standby database runs in recovery mode until there is a failure at the primary site. At the time of a failure, the standby database performs recovery operations and comes online as the primary database.

A standby database uses the archived log information from the primary database, so it is ready to perform recovery and go online at any time. When the primary database archives its redo logs, the logs must be transferred to the remote site and applied to the standby database. The standby database is therefore always behind the primary database in time and transaction history.

The physical hardware on which the standby database resides should be used only as a disaster recovery system; no other applications should run on it. Because the standby database is designed for disaster recovery, it ideally resides in a separate physical location than the primary. The standby database exists not only to guard against power failures and hardware failures, but also to protect your data in the event of a physical disaster such as a fire or an earthquake.