The print spooler daemon acts as a sort of pipeline. It accepts print jobs coming in from various print clients, and then passes these jobs along to the appropriate printer. While the printer is busy, these jobs "spool," waiting for their chance to get printed.

When printing on the local printer, both of ends of this "pipeline" are described by the configuration file /etc/printcap (sometimes located at /etc/lprng/printcap). Each entry in the printcap (which is short for printer capabilities) describes one print spool:

$ more /etc/printcap
lp|Generic dot-matrix printer entry:\
        :lp=/dev/lp0:\
        :sd=/var/spool/lpd/lp:\
        :pl#66:\
        :pw#80:\
        :pc#150:\
        :mx#0:\
        :sh:

Note that the last line of the entry does not have a trailing backslash (\).

Your distribution may have other entries, and they may be more complex, but they should all have roughly this form. The name of this entry comes first, lp, followed by a longer description of this spool. The keyword/value pair lp=/dev/lp0 specifies the Linux device were print jobs in this spool will be printed, and the sd keyword gives the directory where jobs will be held until they can be printed.

The rest of the keyword/value pairs provide details about what type of printer is hooked up to /dev/lp0. They are described in the printcap man page, and we'll cover some of them a little bit later.