We’re going to start looking at Clustered Print Servers. Clustering itself tends to intimidate many administrators, especially newer administrators. Much of the fear factor usually stems from being unfamiliar with how clustering works, and the fact that you are dealing with shared storage. Throw in the fact that most production clusters are providing mission-critical services and it’s easy to see how an administrator might be somewhat hesitant regarding clustering. While we’re not really going to cover clustering itself in this post, we will touch on some basic concepts as we go along. So let’s get started …
So what exactly is a print cluster? A print cluster is a print server that is configured as a spooler resource on clustered servers. In a Windows Server 2003 cluster server configuration is a group of independent computers that work together to provide a common set of services and present a single-system image to clients. What this means is you group two or more physical servers presenting a single virtual server name to clients. That virtual server name is the access points to services created on the cluster, including printing. For example: You have 2 physical servers, NodeA and NodeB. Those servers are in a cluster group and have a spooler resource installed. That spooler resource has a name associated with it: PrintServerA. Clients access ‘\\PrintServerA’ to gain access to printers installed on the spooler resource.
There are numerous services and applications that can be clustered – for example, SQL and Exchange are often clustered. Network services such as File Servers and Print Servers are perhaps the most common examples of non-application clusters. We are going to focus on Print clusters since that is primarily the area that the Performance team supports. When you think of a print cluster, just remember that what you are dealing with is a print server that is installed on a clustered set of servers and presents a single virtual server name to clients. The print cluster provides a high availability solution for printers. Thus, if the active node (server that owns the print resource) fails, the print resource fails over to the next server in the cluster. This allows the printers to remain available by maintaining the same access point for clients. So in our example above, if NodeA owns the print resources and fails for some reason, then the resources would fail over to NodeB. From the client perspective, they would still just access the virtual server name – \\PrintServerA.
- KB 278007: Available features in Windows Server 2003 Clusters
- Technet: Checklist for creating a server print cluster
- Creating / Configuring a Highly Available Print Server on Windows Server 2003
- Technet: Additional Resources for Designing & Deploying Print Servers
Microsoft, Windows 2003, Print Clusters, Printing, Trobuleshooting, Knowledgebase