System Freezing

[Douglas K. Rand]

I'm having what is probably a hardware problem on a system that just
hangs every 6-36 hours, and I'm wondering if anybody has any ideas for
things I could try.

Its a RELENG_4_8 system with DDB, DDB_UNATTENDED, and ALT_BREAK_TO_DEBUGGER
kernel options set. (Its on a serial console, thats why the
ALT_BREAK_TO_DEBUGGER option.) Its an Athlon 3200+ on a Gigabyte
GA-7N400-L mobo, with two 512MB PC3200 DDR DIMMs, and a 2 port 3ware
controller and 2 Deskstar 180 GXP disks. The power supply is an Antec
TruePower 380W.

The system ran perfectly for about 60 days, and then started having
this problem. In almost all cases the system will simply hang, there
is no response from the console or network, and the CR ~ ^B sequence
will not get me to the kernel debugger. (I've tested this when the
system is running fine and I do get the kernel debugger.) The only
solution is to reset or power cycle the system.

It has crashed 3 times with a Fatal trap 12: page fault while in
kernel mode panic, and one time it simply rebooted as if someone
pressed the reset button. But it has simply hung 18 times.

I've tried running with only one DIMM, and when the system died 3
times with that DIMM, I tried running with only the other DIMM, and it
still dies.

I've replaced the power supply with an Antec 400W, and the system
still dies. I even replaced the power cord.

I've tried both the stock 4.8 twe driver and 3ware's beta driver, both
still die.

I replaced the onboard NIC with an Intel Etherexpress Pro, and the
system still dies.

I don't think its temperature related, I've run the system with the
case open and on its side, and a continous mbmon output shows no
temperature increases just before the system hangs. (A representative
output from mbmon is:
  Temp.= 75.2, 113.0, 86.0; Rot.= 4821, 2636,    0
  Vcore = 1.70, 2.74; Volt. = 3.31, 4.14, 11.55,  -5.29, -2.05
I've got a ThermalTake Volcano 11+ cooler on the CPU.

I don't think the problems are load related, as it carries very high
loads with out hanging, and I've had it hang with fairly light loads.

I've attached the dmesg and kernel config files. If anybody has any
suggestions I'd be thrilled. I'm up to replacing either the CPU or the
mobo, neither of which I'm looking forward too. 

-------------- next part --------------
A non-text attachment was scrubbed...
Name: dmesg
Type: application/octet-stream
Size: 3124 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-hardware/attachments/20040318/29e26123/dmesg.obj
-------------- next part --------------

-------------- next part --------------
A non-text attachment was scrubbed...
Name: SNOW
Type: application/octet-stream
Size: 883 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-hardware/attachments/20040318/29e26123/SNOW.obj
-------------- next part --------------