Mouse doesn't get turned off on system shutdown

[Brian Burger]

Binary package hint: xserver-xorg-input-mouse

Since re-installing 6.06, my mouse has not been powered off on shutdown.

System->Quit->Shut Down shuts the system off normally, except that my optical mouse light stays on until I physically cycle the power switch at the back of the desktop.

It's a basic Microsoft optical wheel mouse, a few years old, and it's been working perfectly until installation of 6.06, so I don't think it's a hardware issue w/ the mouse.

In Windows XP, 5.04 & 5.10, with exactly the same hardware, the mouse always shut down w/ the rest of the system.

[Brian Burger (bburger)]

Just found a thread on ubuntuforums that confirms that this isn't just my hardware, and that it's a problem that seems to have cropped up with Dapper...

http://ubuntuforums.org/showthread.php?t=206385

From some of the comments in that thread, it seems to be a general USB problem, not merely a mouse issue.

[Gerry Neary (gerry-neary)]

With me it's not just the mouse but the whole system. It gets to "will now halt LVM Volume Group" and stops. After awhile it goes into command screen mode and just sits there until I turn the power off. Anything I'm doing wrong as I am newbie

[Liam Legionnaire (malaphar)]

Same with me. All USB peripherals stay powered (mouse, keyboard, drawing tablet, usb stick, etc.)
I really would like to see this fixed, not just for the annoyance, but to conserve the energy.

[Gerry Neary (gerry-neary)]

I found the answer, well in my case at least. I had a couple of dead USB
sticks (USB 1.1) attached to the machine. When I removed them everything
worked fine, Might be worth a check.

I am nor sure it was cos they were dead or cos they were 1.1.

On Sat, 2006-08-19 at 21:25 +0000, Liam Legionnaire wrote:
> Same with me. All USB peripherals stay powered (mouse, keyboard, drawing tablet, usb stick, etc.)
> I really would like to see this fixed, not just for the annoyance, but to conserve the energy.
>
> ** Description changed:
>
> Binary package hint: xserver-xorg-input-mouse
>
> Since re-installing 6.06, my mouse has not been powered off on shutdown.
>
> System->Quit->Shut Down shuts the system off normally, except that my
> optical mouse light stays on until I physically cycle the power switch
> at the back of the desktop.
>
> It's a basic Microsoft optical wheel mouse, a few years old, and it's
> been working perfectly until installation of 6.06, so I don't think it's
> a hardware issue w/ the mouse.
>
> - In XP, 5.04 & 5.10, with exactly the same hardware, the mouse always
> - shut down w/ the rest of the system.
> + In Windows XP, 5.04 & 5.10, with exactly the same hardware, the mouse
> + always shut down w/ the rest of the system.
>

[Tollef Fog Heen (tfheen)]

This sounds more like a kernel problem than an X mouse driver problem -- reassigning.

[Robert Pickle (rpickle)]

I can confirm on my end. This has been the case ever since I started using Ubuntu though, from Dapper flight 4 or so. Pretty annoying.

[Haiyadragon (rl-bekkema)]

I have this will all Linux distros. Any other way of shutting down will cut the power properly.

[Chris Coulson (chrisccoulson)]

I'm also affected by this. I'm running Ubuntu 6.06 with kernel 2.6.15-26-amd64-k8, but can confirm it also happens with 2.6.15-25-amd64-generic, 2.6.15-26-amd64-generic and 2.6.15-27-amd64-k8. I've only ever had the problem since installing Dapper. My USB devices powered down correctly when shutting down from Breezy. Other OS's on my system work fine as well.

[Federico Belvisi (armalite)]

This problem came up to me when I recently upgraded to an nforce4-based motherboard.

I'm using Edgy.

[Bornakke (bornakke)]

Same problem here. Got it with my Web cam, Wireless adapter and mouse. Kernel 2.6.17.10, Ubuntu Edgy, and no problems with XP

[Brian Burger (bburger)]

Just updated (finally) to 6.10/Edgy, and still have this bug - USB systems are still not being powered down properly when the machine is shut down.

Running "2.6.17-11-generic" kernel.

[Sebastian Kemper (sebastian-ml)]

Hi,

I opened a bug report at the kernel bugzilla regarding this one (I think) some time ago. I also found the commit that introduced this behavior.

Link: http://bugzilla.kernel.org/show_bug.cgi?id=5410

Regards
Sebastian

[Giorgio Vazzana (mywing)]

I'd like to confirm that this bug also occurs on my sistem since I've upgraded to Feisty. Basically my optical mouse light stays on after I shut down. The sad fact is that it used to work fine with Dapper, and plus using Dapper live cd it still works fine. Some info about my sistem: Asus K8U-X motherboard, PS2 Mouse, Ubuntu Feisty amd64, kernel 2.6.20-15-generic

[Piotr Adamowicz (citricacid)]

Same problem. All USB devices are powered on after system shutdown. On Windows shutdowns everything turns off.
Motherboard: Epox on nForce2 chipset, Ubuntu 7.04, kernel 2.6.20-16-generic

[Giles Weaver (gweaver)]

Same problem. All USB devices (mouse, external hard disk etc) are powered on after system shutdown. Windows shutdown turns all USB devices off.
Motherboard: Epox on nForce4 chipset, Ubuntu 7.04 i386, kernel 2.6.20-16-generic

Hopefully this will be fixed in a timely fashion with a new Kernel...

[Pascal De Vuyst (pascal-devuyst)]

Brian,
Thanks for your report, more information is needed.

Please provide the minimal information as described here: https://wiki.ubuntu.com/KernelTeamBugPolicies.

Then open a terminal and enter the following command:
  $ lsusb -v > ~/lsusb-v
Attach ~/lsusb-v and /var/log/Xorg.0.log to the bug report.

[Brian Burger (bburger)]

Pascal,

Thanks for adopting this bug - it's been open far too long.

I'm away from home travelling until late in September, but will run
the reports you've asked for as soon as I get back to my home machine.

Thanks!

[Giorgio Vazzana (mywing)]

Pascal,

these are the files you needed. I've used the following script to create them:

#!/bin/bash
mkdir bugreport
cd bugreport
uname -a > uname-a.log
cat /proc/version_signature > version.log
dmesg > dmesg.log
sudo lspci -vvnn > lspci-vvnn.log
sudo lsusb -v > lsusb-v
cp /var/log/Xorg.0.log .

Let me know if you need any other information. Thanks.

[MoMaT (momat)]

Same behaviour on my xubuntu machine.

$ uname -a
Linux ubuntu 2.6.20-16-generic #2 SMP Fri Aug 31 00:55:27 UTC 2007 i686 GNU/Linux
$ cat /proc/version_signature
Ubuntu 2.6.20-16.31-generic

[Brian Burger (bburger)]

Short reports inline here, longer ones attached.

uname -a:
Linux warbard 2.6.20-16-generic #2 SMP Fri Aug 31 00:55:27 UTC 2007 i686 GNU/Linux

version:
Ubuntu 2.6.20-16.31-generic

[Brian Burger (bburger)]

Last of the requested reports from my machine for this bug.

[Sebastian Kemper (sebastian-ml)]

I got this from Alan Stern as a workaround. I hope it works for you guys as well.

It probably won't make it into the vanilla kernel. Read the comments at http://bugzilla.kernel.org/show_bug.cgi?id=5410 if you're interested.

Regards
Sebastian

[MoMaT (momat)]

The bug on kernel.org has been rejected with strange "insufficient data" status, despite a workaround patch being given (using USB controller suspend instead of shutdown). Does this mean that compilation of my own kernel is the only way to solve the problem?

[Federico Belvisi (armalite)]

I added linux-source-2.6.22 since this problem also affects Gutsy.

I tried the workaround Sebastian Kemper posted and it worked for me on a nvidia nforce2 intel based motherboard. Can't this patch find its way into ubuntu kernel? What are the problems other unaffected users can have using this workaround?

[Leann Ogasawara (leannogasawara)]

First, Sebastian, thank you for submitting the upstream bugzilla report. That is a great way to leverage the help and support of the mainline kernel community and often results in a fix that will then get merged with the Ubuntu kernel.

Unfortunately in this case (from reading the bugzilla report) this bug is more likely the result of a faulty BIOS and not the kernel. As Sebastian pointed out, it is not likely the patch from Alan will make it into the mainline kernel. I'd also imagine it is unlikely that the Ubuntu kernel team would maintain this patch locally as it's uncertain how this change would affect other users as well as the added difficulties that come with maintaining out of tree patches. However, I'll reassign and defer to the kernel team for their final say in the matter. Thanks.

[Sebastian Kemper (sebastian-ml)]

Hi!

WRT comment #30: I closed the bug tagged as "insufficient data" because I think in the end in comes down to BIOS intrinsics like Leann says. As the BIOS is proprietary there is "insufficient data" the way I see it :)

Maybe this bug is nforce2 specific. If so we could "lobby" for a kernel quirk. Anyone of you having this problem with a chip other than nforce2?

Regards
Sebastian

[Giles Weaver (gweaver)]

Yep, same thing happens on my Nforce4 motherboard. This bug may be a result of a faulty BIOS, but my guess is that a very large proportion of BIOSes are faulty!

If a kernel workaround can solve the problem I think it should be done. After all if Microsoft can shutdown USB devices properly, Linux should be able to do the same.

[Federico Belvisi (armalite)]

If only i could upload the patched kernel source to ppa, there will be a repository with a kernel for us. But i am still unable to upload, for some reasons it fails (but that's another story). If some of you could try doing this, it will be a really good solution for our "lobby" :)

I wish to thank you Sebastian, you did a real good job here and on the kernel list!

[MoMaT (momat)]

I have this problem with Pentium III motherboard with Intel 440BX chipset. More details in Comment #24.

I was not satisfied with "insufficient data" status because it's rather suggesting that reporter didn't provide enough data to triage the bug, which is obviously no the case here. I would say it should be "testing needed", since more tests are needed to find out what hardware is affected and how the change from USB controller suspend to shutdown will influence previously unaffected machines.

[Federico Belvisi (armalite)]

I finally managed to compile and upload to my ppa a kernel with the patch submitted by Sebastian, you can find it here:

deb http://ppa.launchpad.net/armalite/ubuntu gutsy main
deb-src http://ppa.launchpad.net/armalite/ubuntu gutsy main

I have a little problem in versioning since synaptic shows me after upgrade that the kernel is still upgradeable with the same version. This should not happen if you download and install manually the .deb, or if you compile from sources.
I'll try my best to keep my ppa in sync with the standard kernel for future versions.
Works out of the box for me with the linux-restricted provided by ubuntu so i didn't have to tweak with nvidia closed drivers. You will/can find other packages apart from kernel on this repository, so pay attention if you want to upgrade only the kernel.
No warranty whatsoever :)

[Ben Collins (ben-collins)]

I believe the comments say it all. This has been labeled a BIOS issue, and the current patch to work around it is not correct at all. I'm not sure if a quirk style patch can even be accepted, unless it can be proven that "fixed" BIOS's (if they ever happen) will not suffer adverse affects from the quirk being enabled.

[FilmAficionado (hd-flavorteam)]

I hope this hasn't been said before:

I found this really interesting. So neither my mouse nor my keyboard lights turn off when I "shut down the computer..."

BUT they both turn off (correctly; as they should) when I place my computer on "standby..."

Interesting, or not worth mentioning? This is on a Gutsy install.

[Brian Burger (bburger)]

Just a quick update: this continues to be a bug on a fresh install of 7.04 for me... exact same symptoms & hardware as before w/ previous versions of Ubuntu.

[Piotr Adamowicz (citricacid)]

Well... this bug is still present in 8.04 with Linux 2.6.24-17-generic.

[Brian Burger (bburger)]

No edit function on previous comments here. Gah. I meant 8.04 as well, of course.

Must not post to bug reports before second round of coffee in the morning...