|
|
Management of resources means starting, stopping and monitoring them. In some cases there are dependencies between resources. For example to unmount a file system, the application that uses the file system has to be taken offline before unmounting the file system. To simplify the management of a set of resources and resource dependencies, you can define a service group. You can distinguish between three different types of service groups:
More information about service groups is available at:
http://www.veritas.com
Managing resources depends on the type of the resource. VCS comes with a bundle of resource types, for example Application, DNS, Mount, or Volume. As you add a resource to your service group, you choose a resource type and select whether it is critical or just enabled. Critical means, a failure of this resource initiates a failover of the service group. Enabled means that this resource is monitored by an agent.
Agents are management processes for predefined resource types or third party applications, for example, IBM WebSphere MQ or IBM DB2 UDB. They consist of type declaration files and binaries to enlarge the management capabilities of VCS.
The concept of service groups allows VCS to distinguish between application and node failures. Service groups are switched between the nodes in case of a failover. Therefore, it is necessary to use a Virtual IP Address (VIP), which is also moved when failing over. The use of virtual IP addresses is also known as IP Aliasing.
