Clustering supports the use of a cluster administrative domain to maintain a consistent operational environment across nodes in a cluster.
Applications often require specific system settings or other environmental conditions collectively known as an operational environment. This may include configuration parameters or data, user profiles, job descriptions, as well as system values, network attributes, system environment variables, and subsystem descriptions. Within a high availability environment, the operational environment must be the same on every node where an application can run or store its data.
The resources identified to a cluster administrative domain, called monitored resource entries (MREs), are also identified in an associated peer CRG. The cluster administrative domain monitors the resources for changes and synchronizes any changes across the active domain. Once the domain is created, normal CRG functions are used to manage it.
Each node can be defined in only one cluster administrative domain within the cluster.
Figure 8.Cluster administration domain example
Figure 8 shows an example of a four-node cluster with a cluster administrative domain. Each node is an LPAR, with two LPARs in each system. LPAR 1 is the normal production system. The node roles shown are those of the peer CRG associated with the domain. Node roles for application CRGs and data CRGs used in the normal production environment are not shown.