[ ... ]
First ting I would do is monitor the cpu temp with lmsensors and see if
there is a correlation between cpu temp and the failures.
Yes, probably a good idea. I've never run lmsensors on this box, though, and it doesn't work directly ;-( It starts, but the values returned just don't make sense at all. This is on an Abit KT7 mainboard. Ideas?
- Toralf
Been awhile but seems like I had to tweek some stuff to get it working
right on the boxes I installed it on. Have you checked the docs?
I've checked it briefly, and it looks like the default config is supposed to support this mainboard, but like I said, it doesn't really. I haven't experimented a lot with the setup, though. I have lots of other things on my mind, and since the machine works OK for *nearly* everything I need to do... Also the temperature looks all right when I read it from BIOS setup just after a crash has occured, so it's not *that* likely to be the problem. Finally, I've now actually ordered a different (low-cost) CPU that I will try using as replacement, and I'll get hold of a larger PSU, too, I think.
I've now replaced the CPU *and* PSU, and everything looks very promising. At least, I can run lots of operations that caused crashes earlier - but of course, the relative load will be somewhat loader as this is a faster CPU.
I also tried replacing the PSU only, but after I did that, the kernel crashed directly after being loaded - which struck me as a bit odd.
- Toralf
-- redhat-list mailing list unsubscribe mailto:redhat-list-request@xxxxxxxxxx?subject=unsubscribe https://www.redhat.com/mailman/listinfo/redhat-list
