[18704] in Athena Bugs
linux 8.4.19: kudzu
daemon@ATHENA.MIT.EDU (Aaron M. Ucko)
Tue Jan 30 13:29:55 2001
Message-Id: <200101301829.NAA05409@roger-and-jessica-rabbit.mit.edu>
To: bugs@MIT.EDU
Date: Tue, 30 Jan 2001 13:29:50 -0500
From: "Aaron M. Ucko" <amu@MIT.EDU>
System name: roger-and-jessica-rabbit.mit.edu
Type and version: i586 8.4.19 (with mkserv)
Display type: XFree86 3.3.6 SVGA
Shell: /bin/sh (/bin/athena/bash?)
Window manager: fvwm2m4
What were you trying to do?
Log in.
What's wrong:
X didn't start after the update because /dev/mouse didn't exist.
For whatever reason, Kudzu removed it just before the reboot:
Jan 30 05:32:36 roger-and-jessica-rabbit kudzu: unlinked /dev/mouse (was linked to ttyS0)
Jan 30 05:32:56 roger-and-jessica-rabbit init: Switching to runlevel: 6
When the machine came back up, Kudzu didn't manage to put the
link back:
Jan 30 05:34:50 roger-and-jessica-rabbit kudzu: failed
Jan 30 05:34:50 roger-and-jessica-rabbit kudzu: Hardware configuration timed out.
Jan 30 05:34:50 roger-and-jessica-rabbit kudzu: Run '/usr/sbin/kudzu' from the command line to re-detect.
When I ran kudzu by hand, it said it found a generic serial
mouse on the first serial port (accurate), and presented a
list of models to choose from. I chose the appropriate entry
(Logitech MouseMan/First Mouse) and everything was happy.
What should have happened:
Kudzu shouldn't have gratuitously blown away /dev/mouse in the
first place. Failing that, it should it least have cached
enough information to guess that the mouse was the same model
as it had been.
Please describe any relevant documentation references:
kudzu(1)
