hardware issues - reboots, crashes

[hunky]

I'm running ~amd64 and have been fine for some time. Lately though I've gotten some hard crashes. One morning I found the computer locked up and the speakers were making a cluck sound repeatedly, maybe a couple per second. Had to do a hard reset. When I did the reset, before the BIOS loaded, I could hear the CD player (empty) making some engagement noises repeatedly but the BIOS wouldn't come up. Resets wouldn't work until I unplugged the computer. Eventually I got it to boot but the BIOS loaded very slowly.

This happened a couple times within a few minutes. Seemed to be working then the next day the same thing. I took out my PCI sound card, re-seated the 4 GSKill memory modules (4GB each), and tried removing the memory modules and replacing them one at a time and different orders to try to get the machine to boot. Finally did get it to boot with all memory intact. Now it has been running fine for a few days.

Today however it locked up as I had VLC running. The speakers froze on a syllable of speech - repeating it maybe 2 or 3 times/second. I also had two browsers open with several tabs and a WinXP virtualbox machine running so figured I had stressed it out and after waiting to make sure it wouldn't come back, and trying some keystrokes like Ctrl-Alt-Backspace, I had to do a reset. So after it finished rebooting I only had bash running and started VLC again. Within a minute or so it rebooted itself without locking up. (Plus I'm running the latest portage version of KDE desktop).

There seems to be two different problems. The first problem I couldn't get it to reboot without several attempts. The second problem seems to be maybe something to do with video and it came back normally after it rebooted itself. I don't know.

I went through the docs on troubleshooting hardware - tried stressing the CPU and that checked out fine. It is a AMD Phenom(tm) II X6 1055T Processor. The motherboard is an ASUS M4A89GTD Pro. I changed out the CD player as well when having the first problem so that wasn't an issue.
It has a Corsair 650 watt power supply, which the power supply is always the first thing I suspect. I also looked at the capacitors on the MB and didn't see any signs of swelling or corrosion.

I have a good fan cooling system I think.. don't really have the temps it has been running at but it seems the CPU stress test would have ferreted that out. No other cards.. just onboard video.

There was a red LED light going on and off for the memory when I was having difficulty getting it to reboot. Called the DRAM LED. But it is off now as the computer is running... maybe something to do with pre-POST or something.

So curious if someone has advice. I've looked at the /var/log/messages and dmesg and didn't see anything unusual but I could easily have missed something. the Xorg log didn't show me anything either but I'm no expert.

I could consider upgrading.. keep the case and all but the computer has met my needs (not a gamer but do some CAD stuff and image manipulation.

thanks for looking, JD

[amospalla]

In my experience, the most common causes of hardware problems come from bad memory, motherboard capacitors or power supply.

You can easily discard memory problems with memtest, and motherboard capacitors by opening the case and watching them.

If it is a power supply problem, you may need to replace it to test the system with a good one, which is not that trivial as the other two options.