Windows BSOD is spreading to my linux system
 

Edit: How the hell did I managed to post this in the Laptop and Notebook forum instead of the Hardware forum??? Should I lock this and repost or can this thread be moved to the correct subforum?

Hello all.

I'm having some issues with my windows partition, and lately it's starting to affect my linux system too. I'll describe the entire story, and I was hoping you guys could help me pinpoint what's the issue with my computer.

I've been experiencing a lot of BSOD events in my windows partition lately. It's probably related to my GPU, because they tend to occur while I'm playing games, but sometimes it happens when I'm doing standard stuff, like browsing the web. Once or twice it happened right after reboot.

Also, sometimes after the errors I see a message like "updating backup BIOS image. Writing BIOS image.....". It reboots normally after that.

Then It started to complain about corrupt files: "The file is possibly corrupt. The file header checksum does not match the computed checksum". I did 2 chkdsk /f after that, the first one found and corrected a bunch of errors.

After that (and while I was gathering the windows crash dumps to make a post in a windows 7 BSOD forum), it crashed again. And this time it's refusing to boot on windows. So, I booted on linux, and it crashed a few seconds after login, while it was loading chromium.

After another reboot, it threw me a "Unaligned pointer 0x7fcf4" message even before grub, and I had to reboot again into recovery mode. After a quick fsck (which found and fixed a couple of errors), I was finally able to log in.

I'm using Ubuntu 11.10. Where can I find the crash logs for the linux crash, and what can I do to figure out what's happening to my computer, and which piece of hardware is trying to kill it?

From your error description I would assume problems with either your harddisk or (what I think is more likely) with the RAM. I would recommend to test both, the RAM with Memtest86+, the disk with the manufacturer's diagnosis tool.

Offtopic: As you have requested I have moved your thread to the Hardware section. For future reference, if you want to have one of your threads moved just hit the "Report" button and ask for moving the thread.

Additionally, you should check the SMART status of the hdd - the "Disk utility" tool has it built-in. See if it reports bad sectors.
Anyway, the memtest check should come first.

Run the memory tests for a few hours or more and then move on to the factory or oem hard drive diags.

Anything from motherboard caps to psu to video card to radar nearby.

I run memtest86+ and it found a bunch of errors in one of the runs. I then opened the case and found that the RAM slots are located awfully close to the processor, and one of the sticks was dusty as hell (I somehow neglected to clean it the last time). Cleaned both sticks, tested them individually for just 1 pass, both came out clean, tested them together, no errors for 8 or so passes.

I'm running the HD self test from Disk Utility now, and I'll leave the memtest running tonight. If everything is ok, I'll reinstall windows and stress test the system by playing games. Hopefully this entire ordeal was caused by dust on the memory stick, and no permanent damage is done.

Thanks for the help guys.

Dust generally isn't bad from an electrical point but could cause heat. I suspect that you may have also forced the connections to clean up a bit too while doing this.

We have had some dust that was electrically conductive and shorted out boards.

Run memtest on it again in a week or so to double check. Any data on your system is now suspect. I'd reload it myself.