>> Could you send (either to me or to the list) your /etc/fstab and
>> what disklabel reports for your disk partitioning?

> liz# cat fstab
> /dev/sd0a / ffs rw 1 1
> /dev/sd0b none swap sw 0 0

Okay, sd0a is a filesystem and b is swap, no other partitions used....

> partition      start         (c/t/s)      nblks         (c/t/s)  type
> 
>  a (root)     179744     (274/00/00)    1865008    (2843/00/00)  4.2BSD
>  b (swap)        656       (1/00/00)     179743     (273/07/81)* swap
>  c (disk)          0       (0/00/00)    2044752    (3117/00/00)  unknown

Here's your problem: 656+179743 > 179744.

I really wish more people would use sunlabel, or that edlabel/disklabel
would acquire a graphic partition display.  (I will gladly give the
sunlabel code for the graphic partition display to anyone interested.)
When I put your disk's numbers in a "label" with sunlabel and ask it
for a partition display, I get this (skipping start=0 size=0 lines):

a: start cyl =    274, size =  1865008 (2843/0/0 - 910.648Mb)
b: start cyl =      1, size =   179743 (273/7/81 - 87.7651Mb)
c: start cyl =      0, size =  2044752 (3117/0/0 - 998.414Mb)

      aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
 bbbbbb
cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc

As you can see, the end of your b partition overlaps the beginning of
your a partition (by 0/7/81, to be precise).

This is not a problem wrt swapping unless swap fills up completely, but
it does explain why kernel coredumps corrupt your root - kernel cores
go at the *end* of swap, to keep them from being overwritten by normal
swapping as long as possible.

I'd suggest you shrink your swap area down to 273/0/0 (179088) and see
if your problems still persist.  I expect the MMU faults to persist,
but I expect the kernel coredump problem to be cured.

I currently suspect flaky hardware for the MMU fault problem, but I do
not by any means consider myself an expert on the subject.  Having a
kernel core to poke at might help.

					der Mouse

			       mouse@rodents.montreal.qc.ca
		     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B