Information on Groups & Resources Using Cluster Server

• Microsoft Windows NT Server, Enterprise Edition version 4.0 - Microsoft Cluster Server

SUMMARY

Microsoft Cluster Server (MSCS) for Windows NT Server, Enterprise Edition (NTS/E) provides higher availability and easier manageability for important resources and applications. MSCS runs as a system service and monitors the health of resources defined within the cluster. Cluster resources are organized by groups of similar or dependent resources.

MORE INFORMATION

Using MSCS, groups of resources may be defined to develop logical groupings for important resources. Some examples of resource types may be applications, services, disks, file shares, print spoolers, TCP/IP addresses, network names, and so on. Other types of resources within the cluster may be added by software developers through custom DLLs if necessary. Many applications and services may not require special handling and may use generic resource types as provided with the product.

Groups of resources allow similar or dependent resources to be grouped together. For clients to access resources within a group, the group must have a network name and IP address resource associated with it.

These groups of resources may be owned and managed by any member computer (node) in the cluster. In the event of a failure within a group, the cluster software transfers the group of resources and data to a remaining node in the cluster. The network name, address, and other resources for the moved group remain with the group after the transfer. Therefore, clients on the network may still access the same resources by the same network name and IP address -- despite the name of the computer that offers the resources. Thus, these resources become more available. Groups may be moved automatically because of a resource failure within the group, or by an administrator for load balancing or administrative purposes.

Multiple groups may be created within the cluster so that resources may be distributed among available nodes in the cluster. The ability to distribute groups independently allows the workload to be handled by more than one cluster node. Administrators may use each cluster node for normal day to day operations. In the event of a failure, the groups handled by a failing node will be automatically transferred to surviving nodes within the cluster.

Resource dependencies may be established within a group to ensure availability of specific resources before other resources attempt to go online. For example, a file share resource may require a physical disk to be online to provide data to clients.

Another example of resource dependencies may be for a Web server resource that needs an IP address to start. If you create the resource with a dependency on the IP address resource, the web service will not attempt to start until this becomes available.

For more information on groups and resources within MSCS, consult the MSCS Administrator's Guide.