Subject: Re: X from a serial console machine?
To: Matthias Drochner <M.Drochner@fz-juelich.de>
From: Jeff Rizzo <riz@boogers.sf.ca.us>
List: port-i386
Date: 02/25/2000 11:17:32
On Fri, Feb 25, 2000 at 08:00:06PM +0100, Matthias Drochner wrote:
> 
> riz@boogers.sf.ca.us said:
> > when booting with serial console, ttyE0 does not get configured.  (I
> > can't start a getty on it from /etc/ttys) ttyE1 *does* get configured
> > with the default wscons setup
> 
> Have a look at /etc/wscons.conf. ttyE0 is commented out because
> people got nervous about the warnings which got issued if it was
> already set up as console. Removing the comment should do it.
> 

ah hah!  Good to know.  I knew there must be something simple I was missing.

> > ttyE1
> > *does* get configured with the default wscons setup, and I can start
> > a getty on it just fine, but I don't know the magic incantation
> > to get X to use it.
> 
> A "vtNN" argument to the X server should do it - my
> /usr/X11R6/lib/X11/xdm/Xservers contains eg
> :0 local /usr/X11R6/bin/X vt05
> which makes it use ttyE4.
> (Note that X starts counting at 1. Probably written in FORTRAN.)

I had tried this, but of course I tried "vt01", thinking that would
be ttyE1.  This also is good to know.  :)

Thanks much for your help.

Now, all I have to do is hack X so that it doesn't refuse to start
if there's no mouse.  :)   (I had been using tty01 as a mouse
with nothing hooked to it, but now I've got a box that doesn't
*have* tty01... and the ps/2 mouse device doesn't get configured
if a mouse isn't found at boot...

Thanks again.

+j
-- 
Jeff Rizzo                                         http://boogers.sf.ca.us/~riz