A Man and his Penguin

The computer chucked another wobbly today while I was at work. I came home to find it frozen.

I’m beginning to wonder if it’s a hardware issue. The two obvious fingers to point at are the motherboard and the CPU. CPUs don’t generally stuff up unless they’ve overheated, and this one has never overheated. The motherboard should also be in perfect condition - it’s 100% solid caps and designed for extreme overclocking, so it should be able to last for donkey’s years.

I might try manually setting a CPU voltage; at the moment, the motherboard works out an appropriate voltage. It might be slipping downwards?

I’m also thinking of writing a daemon that monitors absolutely every aspect of your computer; all the sensor information, memory and CPU use, kernel messages, absolutely all the logs I can find. Maybe then I’ll be able to work out what the problem is, and in the end I’ll have a great troubleshooting program!

I bought a Noctua CPU cooler NH-C12P and a Corsair 620W modular power supply yesterday (oh, and a 1TB hard disk drive). It took me ages to install the cooler; I’m afraid I was a bit of a nervous nelly, especially the bit where the instructions tell you to take the cover off the heatsink’s base and there was no cover! The best advice I can give you is to install the backing plate and then put the motherboard back in and hook up as much as possible - otherwise you might not be able to reach the CPU power socket or one of the motherboard screws (happened to me).

It was my first time applying thermal paste but it’s all gone very well, and I followed the instructions to the letter in this regard.

The cooler is awesome. I’m running at 3.6 GHz (up from the 3GHz stock- E6850). The temperature at idle is 40 celcius, but with both cores under fire it goes to 55 celcius. With the stock Intel cooler, the CPU would burn at 66 celcius with a 3.3GHz clock. I can barely hear the fan - that Noctua fan is very well engineered and I’m thinking of getting one for the chassis too.

The power supply is also pretty nice. It’s got much more power than I need, but I bought it because it’s modular.

The only problem I had all day was GRUB. I had unplugged all the SATA cables, as you’d expect, and I put them back in different numbered ports. GRUB wouldn’t load. Once I figured out that it wasn’t a BIOS problem, I just booted from the Ubuntu live CD and reinstalled GRUB.

My crashes are gone - I ran the machine overclocked literally all day and night without issues, and all my encoding got done. One instance of Acidrip segfaulted, but only after the ripping job had finished anyway.

Happy? Yes, I’m ecstatic. Next time we’ve got the airconditioning on I’ll try pushing her to 4GHz; it should be a cakewalk with this cooler and motherboard. I have the P5K Premium motherboard - you just tell it what frequency to run the FSB at, and it adjusts everything (including voltage!) for you.

I’m off to rip the third, fourth and fifth seasons of Quantum Leap. See you later.

I think, hopefully, I might have found the source of all my troubles.

I played Sauerbraten today to test whether the freezes were caused by my CPU or GPU overheating. Sauerbraten ran fine, so I decided to do a more scientific test by using Phoronix Test Suite. When looking at the tests available, I thought “Hmm, I’ll try this CPU stress one”.

It went for an hour without incident. So I decided to do “memory-bandwidth”. No problem. So I decided to do “iozone” (I’d noticed that the crashes only occur when there is data being written to disk). Anyway, I ran the first test and the computer crashed. I rebooted and ran it again, and it crashed again. This time I got a dmesg log - the actual crashing process was “kswap”. I rebooted and ran the test yet again, and it crashed yet again.

So I rebooted and turned off the swap. The iozone test didn’t crash the computer nor make it unstable. I’m now encoding videos, with the only notable incident so far being a single program crash (Prism); probably unrelated.

So, it looks like the problem is corruption of the swap. I don’t know if it’s a software problem or a hardware problem (bad sectors?), but so far turning off the swap has stopped the crashes. We’ll see what happens next morning though!