This function is enabled when building a cluster system in combination with ServerView Resource Orchestrator Virtual Edition.
This configuration provides active, standby and spare nodes. Normally, cluster applications are started on the active node. In the event of a failure on the active node, the cluster applications fail over to the standby node.
After this, the spare node is started up using data from the storage, thus keeping the cluster configuration alive and ensuring even higher availability.
Configuration Update Service for SA is a function that automatically retrieves the BMC or iRMC IP address of the spare node and then updates the configurations of shutdown agents on the local node and other cluster nodes when starting the operating system.
Note
Use the same user name and password for BMC or iRMC on every node.
If the PersistentFault attribute of RMS is set to "1," the Fault information is kept even if RMS is started on a normal spare node. (The default value of the PersistentFault attribute is "0.")
When you update the configuration file for the shutdown agent, the updated configuration file is distributed to nodes in which the communication is available. The file is not distributed to nodes in which operation is stopped or the network communication is not available.
In addition, when you start multiple nodes simultaneously, the configuration file for the shutdown agent is updated and distributed on multiple nodes at the same time. In this case, inconsistencies may occur in the information of the configuration file for the shutdown agent stored in each node.
To check that correct information is distributed to all the nodes, execute the following command on any node when all the nodes are activated.
# /opt/SMAW/SMAWsf/bin/sfsacfgupdate -s
When the information that is output by the command is different between nodes, restore the service according to the procedure in "E.6 Restoration."
