Warm standby mode with DB2 HA Feature for LUW and shared disk

High availability for DB2® can be achieved by using IBM® DB2 High Availability Feature for Linux, UNIX, and Windows with shared disk.

Warm standby mode is when the primary database hardware and software components are supported by a standby database server. The standby server is ready to start when the database fails. Clients access the database by using a virtual address or a service IP address that is reserved for DB2.

High availability with DB2 HA Feature for LUW is achieved by implementing a shared storage system to store the database instance and the database. The shared disk is mounted on the active node and is mounted on the passive node only when a failover occurs. By using a shared disk, there is no replication required to keep the passive node in sync with the active node. A cluster manager such as System Automation for Multiplatforms automates the failure detection and the failover to the passive node. Mounting of the shared disk is also handled by the System Automation for Multiplatforms policy.

A common shared disk can create a single point of failure if the shared file system is inaccessible. You must configure the shared file system for high availability to eliminate the single point of failure.

The following diagram shows a primary database server that uses a DB2 service IP address and a shared disk to support data availability on a standby database server.

The diagram is described in the main body of the topic.

There are various methods of achieving file system high availability. To determine what is supported in your existing environment, see the product documentation and the system specifications for your operating system .



Feedback