When a failure occurs in the system being monitored (hereafter called the active system), HA monitor switches to the secondary system (hereafter called the standby system) to continue processing. This functionality is called system switching.
The following section shows examples of hardware configuration and gives an outline of the operation.
The following example shows a one-to-one system configuration consisting of an active system and a standby system.
The configuration consists of active and standby systems connected by a LAN, in which servers provide services, monitoring paths for the systems to monitor each other, and a reset path for indicating a reset when a failure occurs in the active system. A disk storage device is shared between nodes.
When a failure occurs in the active system, HA monitor stops the active system and switches systems. The shared disk will be mounted onto the standby system. The following figure shows an example of the one-to-one system switching configuration.
Figure C-1 Example of the one-to-one system switching configuration
HP-UX HA monitor can control a double LAN adapter configuration in the system. You can define the LAN adapter used by the servers as a redundant model consisting of an active system and standby system. With this configuration, HA monitor can check the state of the LAN at intervals and if the active LAN adapter fails, HA monitor automatically switches to the standby LAN adapter.
The following figure shows an example of the double LAN adapter configuration
Figure C-2 Example of the double LAN adapter configuration