Cluster environment configuration and use

The term cluster has different meanings in different environments. It can mean highly available, high performance, load balancing, grid computing, or some combination of all of these terms.

There are currently several clustering products available for UNIX and Linux, and this section defines those aspects of a clustering environment that need to exist in order for this backup methodology to work correctly. A basic understanding of how your cluster software functions is needed. Cluster software related activities such as the development of application start and stop scripts are not described in this section.

A cluster environment refers to a UNIX or a Linux environment which exhibits the following characteristics:

• Disks are shared between physical workstations, either in an exclusive fashion (only one host has access to the logical disk at any one time) or in a concurrent fashion.
• Disks appear as local disks to the host and not as network resources.
  Important: Mount the file systems locally to the system, not through a LAN-based file share protocol such as network file system (NFS).
• Mount points of local disks are identical on each physical host in the environment (if file system /group1_disk1 fails from NodeA to NodeB, it is mounted on NodeB as /group1_disk1).