From owner-freebsd-hackers  Thu Jun 20 11:08:47 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id LAA28196
          for hackers-outgoing; Thu, 20 Jun 1996 11:08:47 -0700 (PDT)
Received: from rocky.sri.MT.net (rocky.sri.MT.net [204.182.243.10])
          by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id LAA28181
          for <hackers@freebsd.org>; Thu, 20 Jun 1996 11:08:42 -0700 (PDT)
Received: (from nate@localhost) by rocky.sri.MT.net (8.6.12/8.6.12) id MAA10423; Thu, 20 Jun 1996 12:07:18 -0600
Date: Thu, 20 Jun 1996 12:07:18 -0600
From: Nate Williams <nate@sri.MT.net>
Message-Id: <199606201807.MAA10423@rocky.sri.MT.net>
To: "Marty Leisner" <leisner@sdsp.mc.xerox.com>
Cc: hackers@freebsd.org
Subject: Re: panics using Pioneer 624/adaptac 1522
In-Reply-To: <9606201714.AA05479@gnu.mc.xerox.com>
References: <9606201714.AA05479@gnu.mc.xerox.com>
Sender: owner-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> How do I save/find what was printed out in the panic?

In /etc/sysconfig:
....
# Set to the name of the device for kernel crashdumps, or `off' to
# disable any statically configured dumpdev, or NO for no change.
# The device should normally be one of the swap devices specified
# in /etc/fstab.
dumpdev=/dev/wd0s2b

# Set to YES if you want kernel crashdumps to be saved for debugging
savecore=YES

Obviously, you'll want to use a different dumpdev (which should be the
same as whatever you're using for the primary swap in /etc/fstab), but
the results are the same.

nec:# more /etc/fstab
/dev/wd0a               /               ufs             rw              1 1
/dev/wd0s2b             none            swap            sw              0 0
^^^^^^^^^^^                             ^^^^
proc                    /proc           procfs          rw              0 0
/dev/wd0s2e             /usr            ufs             rw              1 1

Then, if you want to enable it for this session type
nec # dumpon /dev/wd0s2b

And a kernel panic will cause it to dump out a core file to your swap
partition.  Note, you'd better have enough room in /var/crash for an
image the same size as your memory + your kernel.

So if you've got a 16MB system, you need 16MB + sizeof(kernel) in
/var/crash.  If your /var partition doesn't have that much space, you
can create /usr/crash and make a symlink from /var/crash to /usr/crash.
(There are lots of other solutions as well, but that's an easy one).

> Another problem is after this my / file system seemed to have major
> problems
>
> (fsck failed, I did an fsck -y and fixed about 30 errors).

Is this under -stable or -current?  If it's under -stable or an older
-current release you *may* need to run fsck multiple times to
completely clean things up.  Also, 'fsck -y' will sometimes 'fix' a FS
so badly that it's unusable.

> After, I was missing 
> 	/usr
> and
> 	/dev/ttyp[0-9] had strange ownership/permissions
> [along with other problems I probably haven't found yet].
> 
> When I tried to rsh into the machine (everything else appeared to be working)
> I got: "not enough ptys"
> 
> So I removed ttyp[0-9], did mknod (looking at anouther freebsd 2.1 system)
> [MAKEDEVICE ttyp0 appear to work), rebooted, and -- the same thing...

MAKEDEVICE isn't a FreeBSD thing, but if you meant ./MAKEDEV ttyp0 that
won't work.

Try doing:
# cd /dev
# ./MAKEDEV pty0

and see if that helps.


Nate