Tom, I agree with Mike, but also suggest two additional tests: - check your memory using the program "memtest86". You boot either off a floppy or a CD into it and its does an exhaustive set of memory tests - looping until you stop it. The kind of error you describe could be a bad section of memory. http://www.memtest86.com/ - check for a correlation between malfunctions and hot environment. Many of today's cheapest computers have Athlon which have lousy CPU heatsink/fans. I had a similar problem to the one you described and it ended up being a CPU problem. Once I replaced the CPU (under warranty! - amusingly, it is the number printed on the fan they use to track the CPU), I haven't had a problem since. Cheers, Doug On Mon, 2004-07-19 at 12:44, Michael Koppelman wrote: > I'd get that up to 1 gig of RAM for sure. It's cheap. It is very > possible that something along the way is memory leaking. > > Michael > > On Jul 19, 2004, at 11:36 AM, Thomas Droege wrote: > > > On the 15th one of the pipelines failed to produce star lists (.cal). > > It > > did make the collated files (.clt) a real mystery since everything > > seemed > > to be OK. > > > > After consultation with Michael, I again ran the photometry step > > breaking > > the .clt files into chunks. Everything ran. Apparently the pipeline > > was > > running out of memory at some stage. > > > > Later it occurred to me that the computer may have "lost" some memory. > > It > > had been running a long time since a reboot. The computer that failed > > was > > my "Wall-Mart" Lindows special. (Long since reloaded with Mandrake > > linux) > > It is the slowest of the 6 computers in the pipeline and has the least > > memory. 256Mbytes. -- ================================================================== Douglas L Welch | Res office/voicemail (905) 525-9140 x23186 Physics & Astronomy | FAX (905) 546-1252 McMaster University | Hamilton, Ontario | Canada L8S 4M1 | E-mail welch@physics.mcmaster.ca ==================================================================
This is a digitally signed message part