Entering content frame

 How a Hot Standby System Behaves when an Error Occurs 

The cluster software constantly monitors the master component of the hot standby system. If the master component fails, the cluster software notifies one of the standby components that it must take over from the master component.

The new standby component adopts the official name (that is used to address the system externally) from the master component.

The standby component is given authorization to write to the log volume and redoes the most recent log entries that have not yet been read, in particular, it redoes all of the transactions that end with a COMMIT. The standby component then switches automatically to ONLINE operational mode.

The hot standby system configuration does not protect the system against errors caused by users or an application. You therefore must make regular data and log backups in your hot standby system to protect it against these types of error.

If an error does occur, you need to import the data backups and log backups from a specific point in time in the past. For more information refer to the Database Manager GUI: Restoring Data documentation and the Database Manager CLI: Saving and Recovering Database Instances documentation.

 

Leaving content frame