Reconfiguring a cluster dynamically
When you configure a PowerHA® SystemMirror® cluster, configuration data is stored in PowerHA SystemMirror-specific object classes in the Configuration Database (ODM). The AIX® ODM object classes are stored in the default system configuration directory (DCD), /etc/es/objrepos.
You can make certain changes to both the cluster topology and to the cluster resources while the cluster is running. This is called a dynamic reconfiguration (DARE). You can make a combination of resource and topology changes via one dynamic reconfiguration operation.
If you have dependent resource groups in the cluster, see Reconfiguring resources in clusters with dependent resource groups for information on making dynamic reconfiguration changes to the cluster topology.
At cluster startup, PowerHA SystemMirror copies PowerHA SystemMirror-specific ODM classes into a separate directory called the Active Configuration Directory (ACD). While a cluster is running, the PowerHA SystemMirror daemons, scripts, and utilities reference the Configuration Database data stored in the active configuration directory (ACD) in the PowerHA SystemMirror Configuration Database.
If you synchronize the cluster topology or cluster resources definition while the Cluster Manager is running on the local node, this action triggers a dynamic reconfiguration event. In a dynamic reconfiguration event, the PowerHA SystemMirror Configuration Database data in the Default Configuration Directories (DCDs) on all cluster nodes is updated and the PowerHA SystemMirror Configuration Database data in the ACD is overwritten with the new configuration data. The PowerHA SystemMirror daemons are refreshed so that the new configuration becomes the currently active configuration.
The dynamic reconfiguration operation (that changes both resources and topology) progresses in the following order:
- Releases any resources affected by the reconfiguration
- Reconfigures the topology
- Acquires and reacquires any resources affected by the reconfiguration operation.
- All nodes are up and running the AIX operating system and able to communicate with each other.
- Any change you make to a cluster definition must be done on an active node.
- The cluster is stable; no recent event errors or config_too_long messages exist.