[Discuss] The T61p meets Kubuntu
Alan W. Irwin
irwin at beluga.phys.uvic.ca
Wed Nov 7 16:17:09 PST 2007
On 2007-11-07 11:53-0800 Michael wrote:
>
>
>> BTW, you might be able to approximate that sane method of installation by
>> using ctrl-alt-F1 to get access to the first Linux text console, edit
>> xorg.conf, then get back into X with alt-F7, then ctrl-alt-backspace there
>> to exit X in the normal way (which gdm|kdm|xdm immediately and automatically
>> restarts, but with the new xorg.conf).
>>
>> Anyhow, it is that old-fashioned "startx" method I always use with my Debian
>> installs, and I am glad I continued that habit with my recent install,
>> because X configuration was tough on Debian as well (more later).
>>
>> Alan
>
> That's actually one of the first things suggested. Problem was, when
> I say brick, I truly mean brick. The keyboard was quite literally dead.
That story is consistent with one of my own recent X configuration
"adventures". When I used
Option "PreferredMode" "1024x768"
to specify that known modeline (as opposed to creating a uniquely named
special modeline I had generated for myself using gtf and specifying that
special modeline with the PreferredMode option) X completely hung somewhere
in the modeline initialization code so none of the usual X key combinations
(e.g., ctrl-alt-backspace or ctrl-alt-F1) would work. However, since I had
initiated X with the startx method, the ordinary console keystroke commands
could be understood such as ctrl-c which allowed me to kill the hung X
initialization and get back to console mode. I suspect in your case even if
ctrl-c actually worked, it would only work for a microsecond until
gdm|kdm|xdm would throw you back into X in the bricked condition again.
Thus, you would perceive your computer as continuously bricked.
My advice to Ubuntu (and all Linux distributions) is to always allow the
startx way of starting X as an option to deal with situations like right now
when X configuration is going through a painful revolution. Then when users
report a bricked condition, cookbook instructions to use the startx option
can always be given to them to work around the problem. This makes sense
not only from the perspective of treating your users right, but also from
the enlightened self-interest point of view where you have much more of a
chance to work through xorg.conf problems for the large variety of your
users' hardware only if you can unbrick them so at least they have a chance
to feed back information to you from the console.
In short, I completely understand the reasons for inventing the gdm|kdm|xdm
display managers to make X startup automated, but there are times when you
don't want that automation so the manual startx method should always be an
option for every Linux distribution.
Alan
__________________________
Alan W. Irwin
Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).
Programming affiliations with the FreeEOS equation-of-state implementation
for stellar interiors (freeeos.sf.net); PLplot scientific plotting software
package (plplot.org); the libLASi project (unifont.org/lasi); the Loads of
Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________
Linux-powered Science
__________________________
More information about the Discuss
mailing list