Using Pocket Print Servers

by Dan Goodell
 Main   Back to Tutorials Index   About Printservers   Configure Printserver   Add Printer 
Configuring the Print Server
This is the trickiest part, and is probably what has given these devices a reputation for being difficult to use. If you're using a network of NT-based computers, I recommend you just follow two simple rules: do not use the manufacturer's software, and set the print server to use a static IP address. With that, you're ready to add the printer to Windows.

To understand the problems, let's first step back and review some basic networking concepts. Most of these devices are quite versatile and may support a number of different networking protocols (TCP/IP, NetBEUI, IPX, AppleTalk) and/or operating systems. However, the following discussion will be limited to what is likely the most prevalent environment: an ethernet network of Windows NT-based machines communicating via TCP/IP.

Devices on a TCP/IP network use IP addresses. No two devices on the same network should use the same IP address. Each device should have its own IP address to distinguish it from the other devices. The exact IP address is not important (and in fact may change from time to time), but in order to communicate with each other devices must use IP addresses in the same subnet. A subnet is a range of IP addresses.

Common subnets used in home LANs are a group of roughly 253 IP addresses covering the range, or, or, for example. If, say, two computers are using IP addresses of and, they will be on the same subnet and able to communicate. In contrast, if the computers are using and, they will not be able to communicate because they are on different subnets--the third dotted numbers are different. (Sticklers may argue that subnets can be tweaked, but that's picayune. The point is that devices must be on the same subnet, and in typical home LANs that usually means the first three dotted numbers of the IP address must be the same for all devices.)

A computer's IP address can be set manually. However, this has a few drawbacks. If you set the IP address manually, you also have to manually set some other things (subnet mask, gateway IP, DNS server IPs). Most important, though, is that it is then up to you to make sure IPs don't conflict. Also, if the machine is a laptop that travels from place to place, you'll probably have to keep changing the settings as you move back and forth from an office LAN to your home LAN.

DHCP (Dynamic Host Configuration Protocol) makes things easier. Rather than manually setting a fixed IP, you can set Windows to "Obtain an IP address automatically." Windows will ask the network what IP it should use, and a DHCP server on the network will give it an IP address. As a laptop moves between office and home, it can automatically get an IP address that works for the LAN to which it is connected.

The common home LAN is based around a broadband router, with the router having a built-in DHCP server function. The DHCP server doles out IP addresses that are in the same subnet as itself, so if all your computers ask the DHCP server for their IPs, they will all be able to communicate with each other. The DHCP server also keeps track so it doesn't give the same IP to two machines. Note that any machine may get a different IP from the last time it booted up if the DHCP server has given the previous IP to someone else in the interim.

If you manually set your computer's IP address, it's up to you make sure it's in the same subnet as everybody else. It's also up to you to make sure it doesn't conflict with anybody else. A DHCP server assigns IP addresses from a pool within the same subnet, so if you manually set your IP it is wise practice to use an address that is in the same subnet but not in the DHCP pool. For example, a Linksys router's defaults are to use for itself and fulfill DHCP requests from the pool If you manually set an IP address, choose something in the range or

last revised: 08/25/2007

Valid HTML5 author: Dan Goodell