System randomly freezes / locks up possibly due to prism54

Ok so it's been two weeks and nothing changed.

There are several forum threads describing similar problems:

http://ubuntuforums.org/showthread.php?t=587905
http://ubuntuforums.org/showthread.php?t=585714

One thread has 11 pages, the other has 22 pages so it seems I'm not alone with this problem to say the least.

Is there anything we can do to help, post back traces, config files.. anything?

So far it seems there is no official comment by an Ubuntu developer.

Let me add my name to this issue.

This is my configuration:
Fresh install of Gutsy
Dell Dimension E521
Dual in-line 1G, 533M, 128X64, 8, 240, 2RX8
AMD Athlon 64 x2 Dual Core 3800+
GeForce 7300 LE

My symptoms are pretty close to the OP, although I don't any particular problems related to my nic. I can be doing anything or nothing, and it will simply freeze. I've had it run an entire day (with me not at the keyboard), I've had it crash overnight (with me not at the keyboard).

There was some suspicion that it was related to my NVidia driver, but I've tried the two latest through Envy and neither brought me joy.

Most of my comments (and others) can be located here:
http://ubuntuforums.org/showthread.php?t=587905

I'm in this as well.

I'm running an upgraded Gutsy (7.04->7.10) which had no stability issues at all before the upgrade.

My problem occurs with the wireless NIC enabled with either bcm43xx+fw or ndiswrapper, or completely disabled.

This does, though, seem to be the kernel and not Ubuntu specific. Slowly more and more threads are appearing with people complaining about this kernel within various distros, and I even found one page where a Gutsy user recompiled his own kernel (disabled SLUB and Tickless mode) and it fixed his problem.

Interestingly, there are things which appear to improve without ever totally fixing the situation:

Worst case: 2.6.22-14-generic with AC unplugged and 'dynamic' power setting.
- Regular lockups, rare to have to battery get to low before the machine's locked up.
Best case: 2.6.22-14-rt with AC connected and 'performance' power setting.
- Has stayed powered up for days on end without fault.

Mine Gutsy i386 on AMD Turion MT-34 and ATI Radeon Mobility X700 is freezing too - up to 50 times per day... I have tried various fglrx drivers, Compiz, xserver-xgl... I have disabled all the extensions in FireFox. Finally, it get a little better when disabling fglrx drivers and using Radeon drivers - but still I am experiencing 1-2 freezes per hour...

I've got what appears to be the same problem. The system in question is a Dell D810 laptop, approx. 2.5 years old, on which a fresh install of ubuntu 7.10 was done shortly after it was released. I'll be working along and suddenly ... nothing. Well, almost nothing: I can still move the cursor around, but the system won't respond at all to clicking on anything usually clickable, and it won't respond to anything typed on the keyboard. For instance, I've tried typing CTL+ALT+DEL and CTL+ALT+F1 after lockup, but there's no response.

Note that this system is dual-boot. It runs Windows XP flawlessly (well, as flawlessly as WIndows ever runs :)) and has been doing so since I got it. So, I have no reason to believe the D810 has a hardware problem.

If anyone has instructions on how to (if it is possible) force a kernel core dump, let me know. Otherwise, I'm going to have to downgrade to ubuntu 7.04 so I can actually do work on this system; or (horrors) go back to Windows.

My /var/log/kern.log contains about 200.000 (yep, that's two hundred thousand) lines that look like this:

Nov 27 01:25:23 localhost kernel: [ 1445.723763] wlan0: TKIP decrypt failed for RX frame from 00:13:49:a1:28:39 (res=-3)

This hardens my initial assumption that it has something to do with wireless.

Does anybody know which package ehis bug should be assigned to? I'm using the prism54 wireless driver, so is this a kernel issue?

As recommended by pedro in #ubuntu bugs I'll attach some info described on the wiki (https://wiki.ubuntu.com/KernelTeamBugPolicies)

After changing my notebook from AMD Turion MT-34 to an Intel Pentium Dual-Core 1.73 GHz - everything's working flawlessly and smooth! I think the freezes were due to AMD CPU - something's wrong with their support - maybe in the kernel...

On Sat, 1 Dec 2007, Aurel Branzeanu wrote:

> After changing my notebook from AMD Turion MT-34 to an Intel Pentium
> Dual-Core 1.73 GHz - everything's working flawlessly and smooth! I think
> the freezes were due to AMD CPU - something's wrong with their support -
> maybe in the kernel...
>
> --
> System randomly freezes / locks up
> https://bugs.launchpad.net/bugs/159398
> You received this bug notification because you are a direct subscriber
> of the bug.

My Dell D810 has an Intel Pentium M 2.13GHz processor. The system is
still freezing at random when running Gutsy. It never freezes when
running Win XP.

This is a follow-up to my previous posting about the freeze/lockup problem with 7.10. I'm the person with a Dell D810 who regularly encounters this problem. Because Tim Fuchs felt this bug might be due to the wireless driver, I hung an ethernet cable down my stairwell and ifconfig-ed my wireless interface down. About 20 minutes later, the system locked up hard, just as it has before. When it locks up, the cursor will move around the screen, but the system is completely unresponsive to mouse clicks or keyboard typing. So, it seems like this bug, or at least the bug my D810 is encountering, is NOT related to the wireless driver.

I guess I should repeat my previous question: does anyone know if there is a way to force ubuntu to drop a kernel core, when the system is locked up such that linux won't respond to anything typed at the keyboard? I'm wondering if there is some way to force it into BIOS, and then if there is something that can be typed at BIOS, that would cause a core to be dropped.

I finally decided to bail out: I replaced the 7.10 install on my Dell D810, with 7.04. I've noticed two benefits:

1) The system doesn't lock up. Under 7.10, it would often lock up (requiring a reboot), most often as a result of heavy browser use.

2) Time to boot the system (from BIOS screen to login screen) has dropped from minutes (didn't measure it) under 7.10, to just 32 seconds under 7.04.

It's pretty clear to me that there is a serious problem with7.10, but not having heard from a developer about this, I couldn't stick it out with 7.10 - it was just too disruptive. So far, for what I do, the functionality of 7.04 is the same as 7.10.

The symptom of this particular bug can be due to many separate issues, so while many people have commented as having the same symptom unless they have the same hardware as the original reporter those should be separate bug reports.

Tim - Thank you for taking the time to report this bug and helping to make Ubuntu better. The Hardy Heron Alpha series was recently released. Alpha2 and subsequent releases contain an updated version of the kernel. You can download and try the new Hardy Heron Alpha release from http://cdimage.ubuntu.com/releases/hardy/ . Please note that Alpha3 will be released within the next day or two so you many want to wait. You should be able to then test the new kernel via the LiveCD. If you can, please verify if this bug still exists or not and report back your results. General information regarding the release can also be found here: http://www.ubuntu.com/testing/ . Thanks!

We are closing this bug report as it lacks the information, described in the previous comments, we need to investigate the problem further. However, please reopen it if you can give us the missing information and feel free to submit bug reports in the future.

I strongly suspect this bug is the same as:

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/204996

If so, the issue is deeper in the kernel because it's affecting multiple WiFi chipsets, multiple video chipsets (Intel, NVidia, etc).

This random lockup issue is just killing some of us on Hardy, and the issues were present from at least Alpha2 that I know and continue right through to the initial production Hardy release even from a clean install.

There's some link to workload. At the moment I'm running release Hardy 32bit via the -rt (realtime) kernel which may handle workload issues just differently enough to escape. Three hours of -rt use so far, no crashes...if it runs all day tomorrow, that will prove this solution to my satisfaction...