Hi Folks, This is a little off-topic, but I thought I'd write a little about a rather bizarre experiment I'm currently running, involving one high-powered 64-bit Linux machine, my currently still cold living room, a central heating system, and a semi-infinitely looped shell script that keeps compiling OpenSceneGraph, and removing the results. Later on this evening, the experiment will probably also include a glass of wine.
Before you start doubting my sanity, here's the background: The high-powered linux machine has been known in the multiplayer community for a brief period under it's call sign PH-UTW, when it logged an exemplary 16 hour+ performance of continual and near flawless FlightGear operations during FSWeekend 2009. This same machine did the same thing after we had assembled it and used it at a public demo that involved a visit from the dean of our faculty. The trouble is that outside of these two occasions in the spotlight, this machine has been plagued with trouble. Right before FSWeekend, I noticed several stability problems, but (thought I) got everything fixed-up just before show time. Right after coming back from FSWeekend, trouble began all over, up to the point where the machine refused to boot. Since then, the motherboard, has been replaced, as have been the two banks of RAM, and the harddisk. In essence, all that is left of PH-UTW are the over sized case, the power supply, the CPU, and the two video cards. And still it keeps crashing. Last weekend I took the machine home, and I started testing today. The hangups are random, but they typically happen during periods of high CPU usage. Compiling OSG gives a good chance of hanging the machine within 15 to 60 minutes after booting it. My current working hypothesis is that these hangups are caused by problems with the cooler, probably due to insufficient cooling grease on the CPU. The exposition hall at Lelystad airport was relatively cold, which could explain why this computer had been running rock-solid during the event Earlier today, I started testing by placing this machine in a relatively cold room, and it has been compiling and removing OSG for 7 hours straight without a single hick-up. So, to test my heat exchange hypothesis, I moved this computer inside my living room, and turned on the heater. If insufficient cooling is a the culprit, the hangups should appear again, once the temperature reaches similar levels it normally is in my office. At this stage, all I can do is wait, and this is where a cool glass of wine enters the equation. :-) To be continued... Cheers, Durk ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
