Cluster resource visibility

  • Article

Applies To: Windows Server 2003, Windows Server 2003 R2, Windows Server 2003 with SP1, Windows Server 2003 with SP2

Cluster resource visibility

When users browse the network, they can see more resources associated with a network name than you might have intended.

This happens because the browsing the network is a two-stage process. First, you locate the computer associated with a network name. Second, you ask that computer to list all of the resources that it owns. If that computer is a cluster node that hosts multiple virtual servers, you can get unintended results. For example, suppose the node called ClusterNode hosts two virtual servers, PrintServer and FileServer. In the PrintServer resource group, you have a network printer named ClusterPrinter, and in the FileServer resource group, you have a file share named ClusterFileShare. When users browse the network, all three network names (ClusterNode, PrintServer, and FileServer) are associated with the same cluster node, and that cluster node owns two resources (ClusterPrinter and ClusterFileShare). Therefore, when users browse the network, they see ClusterPrinter and ClusterFileShare associated with all three network names.

This allows users to circumvent your intended usage of network names. In the previous example, you place PrintServer and ClusterPrinter in the same resource group because you want both the network name and the printer resource to fail over as a unit to other cluster nodes; users who connect to ClusterPrinter through the network name PrintServer can maintain the connection in the event of a failover. However, users in the previous example can connect to ClusterPrinter through the network names ClusterNode and FileServer as well. If users do this, they will lose their connection to ClusterPrinter in the event of a failover because the resource ClusterPrinter will no longer belong to the computer named by ClusterNode or FileServer.

You can minimize the potential for problems by:

  • Using cluster and node names that are easily identifiable by administrators, but not users.

  • Defining virtual server names that are easily identifiable by users.

  • Associating the virtual server names with the resources hosted by the virtual server (for example, ClusterPrintServer is a good name for a virtual server that hosts a printer named ClusterPrinter).

For more information on virtual servers, see Virtual servers.