AMD64x2, Nvidia, Gnome and KDE, no virtual ttys, and shell in konsole or gnome terminal dies

[tiger99]

Binary package hint: xorg

This is possibly not the same as bug 457695 as it is on different graphics hardware, although there are similarities. This is Acer Aspire 7520 with Nvidia graphics and Nvidia driver.

First noticed in Kubuntu 9.10, but parallel installation of Ubuntu on same machine has exactly the same problems, so it is probably not a KDE4 bug.

First indication is that the shell, in a GUI window (konsole or gnome terminal) just dies. It stops responding to keystrokes, but as the GUI is still working, it can be closed. It will then not restart. It dies if a running non-GUI program is generating output, or when manual keyboard input is occurring, and may be a few seconds to (rarely) an hour or more after startup.

On attempting to bring up a virtual TTY to investigate, it was found that keystrokes were having no effect except in any open GUI window such as a text editor, where all works normally. After tghe shell dies, keys are not seen outside the GUI window so can't start any of the virtual consoles by <CTRL><ALT>F1 etc.

On fresh boot, again in Gnome or KDE, before shell has died, <CTRL><ALT>F1 etc bring up black screen, with no cursor, no respones to keyboard, etc. Backlight is still on. <CTRL><ALT>F7 returns to GUI correctly. After shell has died, <CTRL><ALT>F1 etc do nothing, so there is no way of getting a dmesg etc after the event. Maybe I will be able to set up SSH to have a look, but not yet, as I need another machine. At the moment the only thing to do is to reboot via the GUI, logging out or restarting X does not seem to get the shell working again.

I think but can't yet prove that there may be 2 distinct bugs here, one because the virtual TTYs are always dysfunctional, and the other, the shell dying, but it could all be the one bug.

The screen resolution is its native 1440*900, if that is relevant.

This problem was not seen in 9.04, 8.10 or 8.04 on this hardware. The OS is currently unusable for serious work because I need the shell. Suggestions for further tests will be gratefully received.

ProblemType: Bug
Architecture: amd64
Date: Mon Nov 16 20:59:04 2009
DistroRelease: Ubuntu 9.10
InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Release amd64 (20091027)
Lsusb:
 Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
 Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
 Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
 Bus 003 Device 002: ID 046d:c019 Logitech, Inc. Optical Tilt Wheel Mouse
 Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
MachineType: Acer Aspire 7520
NonfreeKernelModules: nvidia
Package: xorg 1:7.4+3ubuntu7
ProcCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.31-14-generic root=UUID=385e93a7-dc23-4854-a505-8c67149510c6 ro quiet splash
ProcEnviron:
 LANG=en_GB.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-14.48-generic
RelatedPackageVersions:
 xserver-xorg 1:7.4+3ubuntu7
 libgl1-mesa-glx 7.6.0-1ubuntu4
 libdrm2 2.4.14-1ubuntu1
 xserver-xorg-video-intel 2:2.9.0-1ubuntu2
 xserver-xorg-video-ati 1:6.12.99+git20090929.7968e1fb-0ubuntu1
SourcePackage: xorg
Uname: Linux 2.6.31-14-generic x86_64
dmi.bios.date: 10/12/2007
dmi.bios.vendor: Acer
dmi.bios.version: V1.09
dmi.board.name: Fuquene
dmi.board.vendor: Acer
dmi.board.version: N/A
dmi.chassis.type: 10
dmi.chassis.vendor: Acer
dmi.chassis.version: N/A
dmi.modalias: dmi:bvnAcer:bvrV1.09:bd10/12/2007:svnAcer:pnAspire7520:pvrV1.09:rvnAcer:rnFuquene:rvrN/A:cvnAcer:ct10:cvrN/A:
dmi.product.name: Aspire 7520
dmi.product.version: V1.09
dmi.sys.vendor: Acer
fglrx: Not loaded
system:
 distro: Ubuntu
 architecture: x86_64kernel: 2.6.31-14-generic

[WeatherGod (ben-v-root)]

tiger99, thank you for your bug report. We are very anxious to understand this problem. When did this behavior start? Could it have started after a particular update? Also, are you using compiz window effects? If so, would turning compiz off stop or reduce the frequency of this issue?

[tiger99 (tiger-sardius)]

WeatherGod,

Thanks for your quick response.

I am not using Compiz, and have now uninstalled it, and I can't yet tell if it started after a particular update.

As I always use separate /home partitions for my data, it should be fairly easy to start again with a fresh install, and add the few drivers and updates, one at a time, until it breaks.

I will report back when I have done that. I will start without the Nvidia driver, as it has to be one of the suspects. Also I will do it on the Gnome installation, to avoid confusion with KDE4 bugs, of which there still are more than a few.

[tiger99 (tiger-sardius)]

Before doing that, I have just tried a desktop of roughly similar configuration, but faster,, i.e. AMD64 dual core and Nvidia graphics, also an up to date Ubuntu 9.10, and that has no virtual consoles either. The monitors show no signal, but <CTRL><ALT>F7 gets back to the GUI.

Beginning to look like an "unfixable" Nvidia problem, but I will proceed to reinstall anyway to see what happens.

[tiger99 (tiger-sardius)]

Well, I reinstalled the laptop, and virtual ttys worked normally. I then used the Hardware Drivers applet, which offered the Broadcomm B43 wireless driver and the Nvidia "185" driver. I have left out the Broadcomm for now, but when I installed the Nvidia and rebooted, virtual consoles were again a dysfunctional black screen. I should have guessed before filing the bug report, but on the other hand it is many years since I have had a Nvidia bug, and that was before Ubuntu existed.

I am going to do the desktop machine as well, just to confirm this. but it will be at least 24 hours.

I will also bring my best machine, a quad core with ATI graphics, up to 9.10 to see what happens there, just in case it is more complex than it seems.

I guess that when I have gathered more data, the only thing that will be possible is to feed the bug report upstream to Nvidia, and hope that they fix it. This is not the place for a rant about closed source drivers, but I have an opinion, and it is not good!

One more thing to do tonight. I have 20 minutes or so, so I will try the Nvidia "173" driver. Don't know if it will work with my hardware, but worth a try.

[tiger99 (tiger-sardius)]

Well, the laptop works with the Nvidia "173" driver as far as virtual consoles are concerned. It may be possible to fix this for everyone who may be affected by making the Hardware Drivers applet offer the 173 driver only, on this hardware configuration.

But give me 48 hours to check the desktops and see what driver, if any, will work there. The fix may be different.

[WeatherGod (ben-v-root)]

tiger99, thanks for the investigative work. It just so happens that someone else with similar trouble was also using the same version of the NVidia drivers. We will see if rolling back solves the issue. It should at least give use a spot to look at to figure out what went wrong.

[tiger99 (tiger-sardius)]

WeatherGod,

OK, just one more report from me. I have checked out the desktop machine too, and the problem is definitely the same, works with the free driver, or the 173, but not the 185. And, lspci reports "nVidia Corporation G73 [GeForce 7600GT] (rev a1)

So it may be only 7 series cards which are affected, but others will be able to confirm that, maybe.

Anyway, I will leave you to get on with fixing it. Just reverting to the 173 driver is probably good enough for many people.

I will keep tracking this bug for a while, so if you need a dmesg or whatever from the other machine, please feel free to ask.

Thanks.

[WeatherGod (ben-v-root)]

tiger99, I am collecting bug reports to send to Nvidia. Could you please (somehow) while using the latest nvidia driver, run the program as root: nvidia-bug-report.sh
This will produce the file nvidia-bug-report.log that you can post to this bug report.

[tiger99 (tiger-sardius)]

WeatherGod,

I have selected a non-working virtual tty to make sure the bug was active, connected via ssh and run nvidia-bug-report.sh as requested.

File attatched. I hope it helps!

[bugbot (bugbot)]

This bug report was filed against an old version of Ubuntu.
Can you confirm whether this is still an issue in natty?

If you don't mind, it would be very helpful if you could update the bug
report in launchpad to 'Fix Released' if it is no longer an issue for
you, or if it is still occurring under natty, please tag the bug 'natty'
so it's easier for us to track.

[mlliso (lisiecki-m)]

I have this problem on natty with NVIDIA so I think it still exists