Red Hat Bugzilla – Bug 18862
Mouseconfig doesn't run when removing USB mouse...
Last modified: 2014-03-16 22:16:41 EDT
First off, I apologise if this is inappropriate, but I've come across
something which seems to be a bug.
I'm running RH 7 on an IBM ThinkPad (2645-9FU). It's a 650MHz Intel
Pentium III, 320MB RAM, 12GB HDD.
If I attach a USB mouse (the optical one from Microsoft), and turn on the
machine, kudzu runs and finds the USB mouse, and I'm given the option to
configure it. If I configure, it installs successfully.
It then asks if I wish to run mouseconfig to update my X information. It
does this successfully, and I can run X/Gnome with no problems.
If I remove the mouse, and restart, kudzu runs, and I can remove the
configuration successfully. However, it does not then ask me to run
mouseconfig! (I was a bit puzzled as to why my X server kept crashing at
first, but when I read the messages generated, it told me that the "core
pointer" wasn't installed.)
This problem is easily fixed by running mouseconfig manually, but it seems
that if it runs mouseconfig when adding a mouse, it should also when it
I haven't tested it on other platforms, or using other mice, but I can't
see the missing call to mouseconfig on removal being related to either my
machine itself or to the mouse, which is why I'm submitting this. I can
replicate the problem consistently.
Or kudzu should at least put the machine back into run-level 3 (after
telling/asking the user), as a configured mouse is a requirement for X.
It could even use the same ugly hack used to move the system into runlevel 5 in
the first place (bug 21061).
*** Bug 20335 has been marked as a duplicate of this bug. ***
*** Bug 29867 has been marked as a duplicate of this bug. ***
With RH71 Beta 4 (Wolverine) running, I added a USB mouse on Dell Latitude
C600. Upon bootup, kudzu failed to identify the mouse as a USB mouse, while in
RH70 (Guninness), kudzu did the right thing. The rest of the story is the same
as in RH70 described above.
Is the USB controller initializing correctly in the case where it
doesn't see the mouse?
Yes, the USB controller is initialized correctly. Kudzu did see a mouse added,
but failed to identify it as a USB type of mouse.
What does the /etc/sysconfig/hwconf look like?
There are two MOUSE entri/etc/sysconfig/hwconf file: one with USB bus for the
added USB mouse and the other one with PSAUX bus for generic PS/2 mouse.
What *exactly* did it say about the mouse that was added?
Here it is:
desc: "Microsoft Corp. IntelliMouse"
If the 'intellimouse' string is what it printed up, that's not a problem;
that's what the mouse is identified as.
Is mouseconfig supposed to come up after the kudzu recognized a new mouse has
been added then? In Wolverine, it failed to invoke mouseconfig at bootup time
while in Guinness it did.
Yes, it *should* come up.
*** Bug 30174 has been marked as a duplicate of this bug. ***
Should be fixed in 0.97.5-1.