The following figure shows a configuration example of the node switching system for the host management model of N executing nodes:
Figure 20-1 Example of system configuration of the node switching system for the host management model
The description of the node switching system for the host management model is as follows:
- This is a cluster configuration system in which N executing nodes and one standby node are deployed on the Application Server machine (host) and Management Server, and Administration Agent is deployed on each node. For every executing node Application Server, there are one or multiple servers with virtual host defined. On the other hand, multiple virtual hosts are defined for one standby node Application Server. For details on the virtual hosts, see 18.3.1 Example of system configuration of the mutual node switching system.
If a failure occurs in the executing node Application Server, the node switches to the standby node. For example, if failure occurs in Management Server of the virtual host 1 of the executing node in figure, the node is switched to the virtual host 1 of the standby node.
- The virtual IP address allocated by the cluster software is used as the IP address for Application Server operations of the executing node and standby node respectively, and the real IP address that does not move to another node by node switching is used as the IP address of Management Server and Administration Agent.
Note that you can execute the following operations in the node switching system for the host management model:
- Using shared disk devices
- A shared disk device is used to inherit the transaction information, such as the OTS status during node switching. Note that if the cluster software is an HA monitor, shared disk devices are not required when using the local transactions.
All Rights Reserved. Copyright (C) 2013, Hitachi, Ltd.