Once during a Linux study at IBM/Amsterdam I had a teacher who made the following statement: "Booting is the hardest part of Linux. But as soon as the kernel has loaded, and init takes over, you're in the clear. From that moment on you read about anything that is happening in the scripts and the man pages."
Off course he's right. There nothing secret about starting up a Linux box. You can read all about it in ASCII files. But it is easy to loose track in all those scripts. And Slackware differs from other distro's because Slackware insists on using a "BSD like" init where other distro's use a "Sys V" init.
Sys V init is a system where all the start/stop scripts are in a directory /etc/rc.d. And for every runlevel there are links to those scripts in a directory for that runlevel (i.e. /etc/rc.1, /etc/rc.2, e.c.t.). There is one big script who traverses the links when entering a runlevel to start (or stop) services for that runlevel.
I will leave describing Sys V init as in use by RedHat, Suse and such to someone with a reference system. I am afraid I would soon get lost in the details without a box actually using it.
Slackers most often have the opinion that a Sys V init is complex and hard to maintain. To be quite honest, RedHat and Suse users think exactly the opposite. But stay with me and judge for yourself.
In this document I will try to prove the IBM teacher right. I will follow the init process of a Slackware (8.1) machine guided by the scripts and man pages. Lots of quotes and little text :D.
Bad english ahead: As you will probably notice, English is not my native tongue. I will try to use a spellchecker as often as possible, but that does not help much with grammar and such. I take no insult in corrections given to me, be it content or grammar. Besides I would like it verry much to hear what people think of my writing.