On Tue, May 29, 2007 at 10:56:40PM +0200, didier derny wrote: > I have some machines running php/mysql/postfix > from time to time the computers running FreeBSD 6,2 > crashes with the following messages > > fatal trap 12:page fault while in kernel mode > cpuid=0; apic id=00 > fault virtual address = 0x104 > fault code = supervisor read, page not present > instruction pointer= 0x20:0xc066c731 > stack pointer=0x28:0xe74fec90 > frame pointer=0x28:0xe74fec9c > code segment=base 0x0,limit 0xfffff,type 0x1b > = DPL 0, pres 1, def32 1, gran 1 > processor eflag=resume,IOPL=0
Is it always the exact same panic or is it random?
A box here had a few random "trap 12 page fault" panics recently,
especially while booting up cold but also at run time, and the
problem turned out to be bad capacitors on the motherboard.
If you have physical access look at the capacitors on the board
to see if they are bulged of even leak.
> cannot dump. no dump device defined
You might want to configure dumping, see dumpon(8).
Look at the traces you get from the dumps:
kgdb /boot/kernel/kernel <dumpfile>
bt
It helps to run a kernel compiled with DEBUG=-g.
I think this is default in GENERIC on 6.2.
If the crashes are at seemingly random points in the kernel
code you probably have a hardware problem. If it's always the
same trace then post it.
--
stefan
http://stsp.name PGP Key: 0xF59D25F0
pgpIIlOG5NjCt.pgp
Description: PGP signature
