The following figure shows an example of system configuration of the N-to-1 recovery system linked with cluster software. In a cluster configuration, if a failure occurs in the running Application Server, the standby node recovery server starts up and terminates the transaction of the executing node in which the failure occurred.

Import all resource adapters that are imported to the executing node, to the J2EE server of the standby node recovery server. Note that you need not deploy applications in the recovery server as the recovery server is only meant to perform recovery for the executing node Application Server in which the failure has occurred.

Figure 19-1 Example of system configuration of the N-to-1 recovery system

You can perform the following operations in the N-to-1 recovery system:

Using a shared disk device

You require a shared disk device to inherit the transaction information, such as the status of OTS. The transaction information is inherited between each executing node Application Server and standby node recovery server.

Applying the load balancer

Though not mentioned in this example of system configuration, you can apply a load balancer by preparing multiple Web servers with the same configuration. You can thereby improve the reliability and operating efficiency of Web servers.

For details on the system configuration of an N-to-1 recovery system, see 3.11.5 Configuration for using a recovery server (N-to-1 recovery system) in the uCosminexus Application Server System Design Guide.