System freezing after login

[deadtom]

I've got an MSI U230 Light netbook running kubuntu 10.10.

This system has been rock solid for months, no problems at all. It was working fine all day today and then just suddenly froze on me. By froze I mean the system simply stopped. Whatever is on the screen is frozen there, keyboard and mouse unresponsive, can no longer access it remotely via ssh, nothing. All I can do is hold the power button until it shuts off and then bring it back up.

Now it does this about a minute after logging in, every single time. I have been able to watch tail -f /var/log/messages and I'm seeing the exact same message appearing every time it freezes:

Feb 18 21:09:41 laure kernel: [ 25.998461] [fglrx] Gart cacheable size:475 M.
Feb 18 21:09:41 laure kernel: [ 25.998468] [fglrx] Reserved FB block: Shared offset:0, size:1000000
Feb 18 21:09:41 laure kernel: [ 25.998471] [fglrx] Reserved FB block: Unshared offset:7ffb000, size:5000
Feb 18 21:09:44 laure kernel: [ 29.170369] ===>rt_ioctl_giwscan. 1(1) BSS returned, data->length = 221
Feb 18 21:09:46 laure kernel: [ 31.174126] EXT4-fs (sda5): re-mounted. Opts: errors=remount-ro,commit=0
Feb 18 21:10:05 laure kernel: [ 49.729084] ===>rt_ioctl_giwscan. 4(4) BSS returned, data->length = 642
Feb 18 21:10:10 laure kernel: [ 54.777170] EXT4-fs (sda5): re-mounted. Opts: errors=remount-ro,commit=0
Feb 18 21:10:14 laure kernel: [ 59.100632] ===>rt_ioctl_giwscan. 2(2) BSS returned, data->length = 364
Feb 18 21:10:14 laure kernel: [ 59.100863] ==>rt_ioctl_siwfreq::SIOCSIWFREQ[cmd=0x8b04] (Channel=1)
Feb 18 21:10:24 laure kernel: [ 69.517907] hda-intel: IRQ timing workaround is activated for card #1. Suggest a bigger bdl_pos_adj.
Feb 18 21:10:35 laure kernel: [ 79.736266] ===>rt_ioctl_giwscan. 3(3) BSS returned, data->length = 506
Feb 18 21:10:47 laure kernel: [ 92.020055] Clocksource tsc unstable (delta = -249766815 ns)
Feb 18 21:11:03 laure kernel: [ 108.495108] [fglrx] IRQ 18 Disabled


The system is freezing immediately after the IRQ 18 Disabled message appears, every time. I don't know what other logs to try to check or where else to look to troubleshoot this.

Google hasn't turned up anything helpful at all.

Thanks!

[Snowhog]

Reboot into a recovery (Single User) kernel so you aren't loading X. Then look at the space stats on your system with:

df -hT

How much is reported as available (Avail) for your root partition?

[deadtom]

Filesystem Type Size Used Avail Use% Mounted on
/dev/sda5 ext4 224G 27G 187G 13% /

[deadtom]

I stand corrected. It just froze again, this time without the IRQ 18 Disabled message.

I don't think it's trying to hibernate or start a screensaver or anything like that because it happens while I'm doing something.

[Snowhog]

Well, okay, that rules out insufficient space on root as a culprit. Prior to this occurring, did you by any chance have a USB Flash Drive or other writable device connected that isn't now?

[deadtom]

Nope, no USB devices were connected.

[deadtom]

I've tried logging in as a different user and it does the same thing. It doesn't seem to do it if I don't log in but I'll let it sit at the log in screen for a while longer and see what happens.

[Snowhog]

In your tail -f /var/log/messages I see:

Feb 18 21:10:47 laure kernel: [ 92.020055] Clocksource tsc unstable (delta = -249766815 ns)

Does this problem present itself if you boot into an earlier kernel?

[deadtom]

And as mysteriously as the problem appeared, it suddenly went away. No idea what caused it or what fixed it.