Thanks for the answers Guys,
I think I got it now.
> The nice thing about the way debian does it, is you can just do a
> /etc/init.d/gdm stop
> (or xdm or kdm, whatever the case may be)
> to stop gdm/kdm/xdm. No switching runlevels and possibly
> starting/stopping other things in the process. Which is nice :)
Actually what you do is edit /etc/inittab and comment out the line
then do,
% telinit Q
Which causes init to re-read the inittab and take the appropriate steps
So you don't change run-levels and as you say risk starting stopping
other things. (But lets not get distracted by that. ;-)
In summary it seems that with the Redhat way you have
init --> xdm --> X
When you kill X, xdm respawns it, when you kill xdm, init (the mother
of all processes) respawns xdm which then respawns X.
Chesty's answer suggests that with Debian you start xdm once from
the rc.n directory, and it spawns X (and respawns X when you kill it),
but if you kill xdm they both stay down. I guess this is fair enough,
I suppose xdm/gdm/kdm aren't very vunerable to crashing on you.
Cheers
Pete
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug