I figured I'd post here first, to see if anybody else has had
similar problems before going too much farther.  I have a 486DX4/100
system that I've been keeping up with current since January.  Basically
I sup the newest sources once a week, and do a `make build` in /usr/src
about every other week, and build a new kernel at the same time.  I've
been having a few problems recently, which I'll list below:

1.  With the "official" 1.2 sources and the last time I sup'ed the
1.2_BETA sources, I've had problems with my machine crashing rather
frequently (once or twice a day, on average).  Originally I thought it
was XFree86, since the instability seemed to come when I updated to
3.1.2F and then G, and I usually run xdm.  However, it seems to crash
even when X isn't running, so I'm starting to wonder if it's a different
problem.
    Given that it's not only under X, about all I can link the crashes
with is dropping a user's priviledges.  The crashes seem to occur with
equal likelihood after either exiting from a logout, exiting from a `su`
or exiting from an xdm session.  From each of these, there seems to be
about a 1 in 4 chance of the machine panicking.  And the panic message
has varied.  Here's a view of my daemon logs:

WARPDRIVE.RES.CMU.EDU (woferry) ~ % grep panic /var/log/daemon | tail -12
Sep 22 03:49:22 WarpDrive savecore: reboot after panic: vrele: ref cnt
Sep 23 03:18:23 WarpDrive savecore: reboot after panic: vrele: ref cnt
Sep 23 21:22:36 WarpDrive savecore: reboot after panic: vrele: ref cnt
Sep 24 02:14:44 WarpDrive savecore: reboot after panic: vref used where
vget required
Sep 24 20:42:07 WarpDrive savecore: reboot after panic: vrele: ref cnt
Sep 25 00:26:04 WarpDrive savecore: reboot after panic: vrele: ref cnt
Sep 25 04:21:04 WarpDrive savecore: reboot after panic: vrele: ref cnt
Sep 25 08:35:14 WarpDrive savecore: reboot after panic: vrele: ref cnt
Sep 26 02:50:50 WarpDrive savecore: reboot after panic: trap
Sep 27 13:19:52 WarpDrive savecore: reboot after panic: vref used where
vget required
Sep 27 16:25:34 WarpDrive savecore: reboot after panic: vref used where
vget required
Sep 27 17:01:40 WarpDrive savecore: reboot after panic: vref used where
vget required

    So you can see how frequently it's crashing (I haven't logged out
since Friday, which is why there haven't been more crashes yesterday and
today...  =), and that the reason for crashing seems to change all the
time.


2.  With the newest X (3.1.2G), has anybody noticed xterm NOT updating
the utmp/wtmp files??  Mine doesn't, and I'm just wondering if anybody
else has seen this.  I guess I'll complain to XFree about that one.


3.  I recently added memory to my machine, going from 32MB to 80MB. 
However, NetBSD seems only to recognize the first 64MB (the kernel
reports 67 million or so bytes available).  I can't find any
documentation on the EXTMEM_SIZE kernel option, is this what I use to
see all 80, and how do I use it?  If I go down to 64, will the 80MB
kernel give me problems?
    Also, is the cache size determined solely based on the amount of
memory?  I've heard that systems like Linux basically use all available
memory as cache, and was wondering how NetBSD handles it.


    I'd greatly appreciate any suggestions anybody might have on these
above problems.  Thanks in advance!

                                                          Will Ferry

--------------------------------------------------------------------------
William O Ferry <woferry@CMU.EDU> | finger:  woferry@WarpDrive.RES.CMU.EDU
talk:  finger for online status   | http://warpdrive.res.cmu.edu/~woferry/
--------------------------------------------------------------------------