Bugzilla will be upgraded to version 5.0 on a still to be determined date in the near future. The original upgrade date has been delayed.
Bug 114591 - anaconda doesn't properly write /etc/sysconfig/mouse
anaconda doesn't properly write /etc/sysconfig/mouse
Product: Red Hat Enterprise Linux 3
Classification: Red Hat
Component: anaconda (Show other bugs)
All Linux
medium Severity medium
: ---
: ---
Assigned To: Jeremy Katz
Depends On:
  Show dependency treegraph
Reported: 2004-01-29 15:19 EST by Joshua Jensen
Modified: 2007-11-30 17:07 EST (History)
2 users (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed: 2004-08-24 11:19:28 EDT
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)
output of kudzu command (3.10 KB, text/plain)
2004-02-18 11:43 EST, Joshua Jensen
no flags Details

  None (edit)
Description Joshua Jensen 2004-01-29 15:19:17 EST
Description of problem:

gpm does not start correctly "out of the box".  The reason for this is
 that after an installed, /etc/sysconfig/mouse does not have a DEVICE=
line in it.  gpm WILL NOT START without that line.  running
mouseconfig, and selecting the same mouse that was selected during
installation will add this line to /etc/sysconfig/mouse... why can't
the installer write this file correctly in the first place?

Version-Release number of selected component (if applicable):

RHEL3 stock

How reproducible:

Do an install (I've got a 3 button usb mouse)
Comment 1 Jeremy Katz 2004-02-12 17:08:19 EST
It works fine without DEVICE=.  See line 40-ish of /etc/init.d/gpm.
Comment 2 Joshua Jensen 2004-02-13 12:22:40 EST
Yet... it doesn't work. :-)   After a kickstart, gpm DOES NOT WORK for
USB mice.  "service gpm restart" doesn't work either.  Add the DEVICE=
line, and restart again, and it works.  Now comment out the DEVICE=
line, restart... and it *doesn't* work.  Noticing a trend here?

/etc/init.d/gpm does say that gpm will assume that DEVICE= /dev/mouse
if something else isn't specified... I notice that on my machines
/dev/mouse is symlinked to psaux.  However, in my kickstart file I
clearly specify "genericwheelusb".  Is anaconda broken in creating
/dev/mouse ?
Comment 3 Jeremy Katz 2004-02-13 13:28:38 EST
It should be getting created via kudzu -- are you disabling kudzu?
Comment 4 Joshua Jensen 2004-02-13 14:33:20 EST
No... kudzu is running every time the machines boots.  It is properly
detecting floppy and cdrom drives... but is stuck on the idea of psaux
for some reason.
Comment 5 Jeremy Katz 2004-02-17 18:49:43 EST
Kickstart or non-kickstart install?  What does `/usr/sbin/kudzu -p -c
MICE` give?
Comment 6 Joshua Jensen 2004-02-18 11:43:34 EST
Created attachment 97804 [details]
output of kudzu command
Comment 7 Joshua Jensen 2004-02-18 11:43:57 EST
kickstart install.  Output of above command is attached as file.
Comment 8 Jeremy Katz 2004-02-18 13:34:09 EST
What sort of mouse line do you have in your ks.cfg?
Comment 9 Joshua Jensen 2004-02-18 17:14:50 EST
Comment 10 Willem Riede 2004-05-02 19:39:17 EDT
I see this problem (Console mouse services fails to start with "no
mouse configured") on my dual PIII test PC (Supermicro P6DBS) after a
clean interactive, custom NFS install of Fedora Core Test 3.

This is with a PS2 wheel mouse behind an ATEN KVM. FC1 and before that
RH8,9 set up the mouse for the console just fine.

Also, anaconda itself recognized my mouse just fine during the
install, and it works in X post-boot.
Comment 11 Féliciano Matias 2004-05-03 07:58:39 EDT
Same here with FC2T3.
Mouse : Generic - 3 Button Mouse (PS/2)

At the first boot, /etc/sysconfig/mouse is empty.
Running system-config-mouse fix this.
Comment 12 Joshua Jensen 2004-08-23 14:21:01 EDT
Any further thoughts about this Jeremy?
Comment 13 Féliciano Matias 2004-08-23 19:42:24 EDT
FC2 (final) and FC3T1 does not have this issue (at least with my

Note You need to log in before you can comment on or make changes to this bug.