I'm going insane. Here's what I'd like to do:

Start KDE on tty7 display :0.0 (already happening
Start the smallest <thing, window manager?> possible on tty8 display :1 (I guess)
Start, full-screen, vmware on the second X - tty8

Currenlty I can:
X :8 -- this brings up what I think I want
vmware -Xqk /home/bloe/vm_machines/xp/Windows\ XP\ Professional.vmx --display :8 -- this brings vmware up to that display, but not full-screen (sure I'm missing a parameter here, may even need to reconfigure it completely since I'm still getting the kernel error message at times).

When I exit vmware X closes down - which is odd - and I don't have anything. Just the blinking cursor in the top left, and I'm unable to change to any other terminal (tty 1-7)!

HELP, before I'm completely bald!!!

tty8 is not setup for login in /etc/inittab for the runlevel you are at. That is why you get the blank screen with a flashing cursor at the top left. However, ctrl-alt-F7 should get you back to your KDE screen, even at the flashing cursor.

I can think of several ways, most hard to explain, to get what you want. I believe the easiest to explain is

1) create a user for vmware,
2) log in on tty2 as user 'vmware' and set up the user's ~/.xinitrc to run vmware as the next to last application in your list (don't forget the &), and a minimal windowmanager as the last item in the list (without the &).
3) add the line DISPLAY=0:1; export DISPLAY in ~/.bashrc (I THINK its 0:1 to refer to tty8).
4) logout and ctrl-alt-F7 to return to KDE.

Start KDE on tty7, ctrl-alt-F2 to tty2, login as vmware, enter the command startx. vmware should come up on tty8. ctrl-alt-F7 and ctrl-alt-F8 to jump between KDE and vmware. When you exit vmware, the minimal window manager should still maintain the X client on tty8. Exit the window manager on tty8 and logout out. ctrl-alt-F7 to get back to KDE.

'course what you are seeing now is the expected functionality of X. The ctrl-alt-F7 should work whether tty8 has a getty line in your bootup init scripts or not.

Kindly note that it would help if you listed your distribution in your profile.

Hey, I didn't put the smilies there. They are supposed to be closing parens.

Yeah, sorry, it's there now. SuSE 10, latest patches and then some.

There has to be an easier way though, isn't there? Couldn't I add the second X in the same script that starts the default one? Running vmware as another user isn't going to work for me, though starting X as a different user shouldn't really matter...I'd simply prefer it to be up when I boot.

As for Ctrl-Alt-<tty>, I know how to do that, but it doesn't work in the situation. The system is still up, and everything is probably running fine, but nothing I do (not even ctrl-alt-del) will get me to a place to even see anything. The only way out is to hit and hold down the power button, then it jumps to tty1 and shutdown the same as it normally would!

Strange. Very strange.

I suppose you could start another instance of X, but in my limited experience, .xinitrc would pose a problem by loading the same apps. See 'man X'.

Where in SuSE does it spawn the original X? Surely you can identify a specific config file for a second one.

Well, I'm not sure about Suse since I switched to Slackware and forgot the SysV init scripts. But I imagine /etc/inittab has a run level of 5 set which starts X and runs kdm for a GUI login.

My minimal understanding of X is that startx (probably startkde on your system) reads ~/.xinitrc to merge in specific X resources and key bindings, and launch your initial GUI applications.

Searching for xinitrc in 'man startx' states that startx first looks for ~/.xinitrc. So, on my Slack system, I entered 'which startx', which pointed me to /usr/X11R6/bin. Reading startx (which is a script), points out real quick that ~/.xinitrc appears traditional (it's not, see further on). It's set in the script, as you suspected. Reading through the script, I find that startx launches xinit (a binary executable). man xinit tells me that xinit launches X.

So, yes, you could. It appears a call to xinit with the proper parameters in ~/.xinitrc would do the trick. See the explanation for

xinit /usr/ucb/rsh fasthost cpupig -display ws:1 -- :1 -a 2 -t 5

in 'man xinit'.

Be very, VERY careful. If your second call to xinit uses ~/.xinitrc, you be in an inifinite loop -- and if your run level is 5, you'll have problems booting. Don't blame me, please!

All my searching has been done using CLI in xterm.