25.1.1 System switchover facility (standby system switchover facility)

If a standby HiRDB is available in addition to the HiRDB that is currently performing work processing and an error occurs on the server machine or HiRDB performing the work processing, the work processing can be switched automatically to the standby HiRDB. The function that switches the system in this manner is called the system switchover facility (standby system switchover facility). Work processing is suspended when an error occurs but before the processing is switched to the standby HiRDB. The purpose of the system switchover facility is to minimize the system down time after an error occurs.

The system switchover facility is used in a cluster system configuration that uses more than one server machine. A HiRDB/Single Server performs system switchover at the system level (note that a utility special unit cannot perform system switchover). A HiRDB/Parallel Server performs system switchover at the unit level.

The system performing the work processing is called the running system, and the system that is standing by is called the standby system. Each time system switchover is performed, the running system becomes the standby system and the standby system becomes the running system. Also, to differentiate between the two systems during system construction and environment setup, the first system to start as the running system is called the primary system and the system that starts initially as the standby system is called the secondary system. Although the running system and the standby system change designations when system switchover occurs, the primary system and secondary system do not ever change those designations. Figure 25-1 provides an overview of the system switchover facility (standby system switchover facility).

Figure 25-1 Overview of the system switchover facility (standby system switchover facility)

[Figure]

1 In this manual, a product that executes system switchover is called cluster software. For details about the cluster software supported by HiRDB, see 25.1.4 Cluster software supported by HiRDB.

2 For details about shared disk units, see 25.5.2 Preparing a shared disk unit.

Explanation
If an error occurs on the running system, the standby system is notified of the error, the systems are switched, and the standby system takes over the work processing as the running system.
Notes about using ClusterPerfect
This manual refers to the running system of ClusterPerfect as the primary system and the secondary system of ClusterPerfect as the standby system.