The complexity of managing potential print service disruption increases as the number of networked computers, printers, and users increases. If you need to replace a printer in an environment in which all networked computers have only local printers and no print server exists, you can maintain users' ability to print by merely replacing the old printer, reloading the drivers, and rebooting the computers.
If an old print server leaves the network and the replacement uses the same network identification and no printers change names, you face only a minor problem. You simply perform a switchover at an off-peak time, rename the old print server with another label, and assign the new print server the old name. Most users will never notice the difference.
You face a major problem if your networked Windows NT print server goes out of service. If the new print server has a different name than the old server, which happens when you leave the older device on the network but reassign it to other tasks, all networked print references point to the old printer resource. Therefore, you must reconfigure the printer setup.
A manual switchover is time-consuming and labor-intensive. Most situations warrant an automated solution. The key to an automated solution is to change the appropriate components. Any scripting language that supports registry calls can automate this process.
The registry on the print server holds printer specifications in the HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Print\Printers subkey. This subkey contains values for printer names, icon names, share names, and port names. The icon name is the label displayed in the users' Printer window, which you access by clicking Start, Settings, Printers.
If you use an automated solution to change these components (i.e., add a new server to the network), the printer name and share name don't have to be the same. However, much confusion results when these two names differ.
—Lee Ratzan
lratzan@scils.rutgers.edu