Customizing the cluster
Customizing the cluster enhances your ability to monitor the cluster and keep it running. You can define a pre-event, a post-event, and a notification method for every cluster event. Notification of events is critical to maintain service for any PowerHA® SystemMirror® cluster. Although PowerHA SystemMirror writes messages to the hacmp.out and cluster.log log files, it is very useful to include notifications to the console or mail to the system administrator when an event occurs that demands immediate attention.
You can include automatic recovery actions as well as notification in the cluster customization. Use the PowerHA SystemMirror and AIX® tools available to customize some or all of the following:
- Hardware error notification
- Hardware failure notification
- Cluster event notification
- Pre-event and post-event recovery actions
- Network failure escalation
- ARP cache refresh
- Pager notification
- Application controller scripts.
It is highly recommended that you maintain a test cluster as well as your production cluster. Before you make any major change to the production cluster, you can test the procedure on the test cluster.