IBM SG24-5131-00 사용자 설명서

© Copyright IBM Corp. 1999

Within an HACMP for AIX cluster, there are several things that are 
customizable. The following paragraphs explain the customizing features for 
events, error notification, network modules and topology services.

An HACMP for AIX cluster environment acts upon a state change with a set 
of predefined cluster events (see 5.1.1, “Predefined Cluster Events” on page 
117). Whenever a state change is detected by the cluster manager, it decides 
which event will be started. It then executes the script for that event in a shell, 
as well as the subevents associated with it. These predefined events can be 
found under 

/usr/sbin/cluster/events. 

The HACMP for AIX software provides an event customization facility that 
allows you to tailor event processing to your site. This facility can be used to 
include the following types of customization:

  • Adding, changing, and removing custom cluster events

  • Pre- and post-event processing

  • Event notification

  • Event recovery and retry

HACMP has the following predefined cluster events:

5.1.1.1  Node Events
This is the sequence of node_up events:

This event occurs when a node joins the cluster. 

Depending on whether the node is local or 
remote, this event initiates either a 
node_up_local or node_up_remote event.

This script acquires the service address (or 

shared address), gets all its owned (or shared) 
resources, and takes the resources. This 
includes making disks available, varying on 
volume groups, mounting file systems, exporting 
file systems, NFS-mounting file systems, and 
varying on concurrent access volumes groups.