Return to BSD News archive
Path: sserve!manuel.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!agate!howland.reston.ans.net!usc!enterpoop.mit.edu!senator-bedfellow.mit.edu!athena.mit.edu!vikki
From: vikki@athena.mit.edu (Vikki King)
Newsgroups: comp.os.386bsd.questions
Subject: *!HELP!* What could be wrong here??
Date: 8 Mar 1993 22:38:50 GMT
Organization: Massachvsetts Institvte of Technology
Lines: 56
Message-ID: <1nghtqINNm21@senator-bedfellow.MIT.EDU>
NNTP-Posting-Host: e40-008-9.mit.edu
Hi Net!
I've run out of remedies - I've got a 386bsd host - a 486 motherboard
from A.I.R, Adaptec 1540B, 1 80MB Quantum SCSI, 1 660MB Maxtor SCSI disk, 16MB
of 60ns RAM and 256k cache. It's also got a multi i/o board w/ IDE (disabled)
and 2 async ports off of a Startek NS16550 compatible controller.
I had this system up running the 0.2.1 patchkit acting as our Internet
gateway via a slip link and a 3com 3C503 ethernet controller. Last night I was
logged in through the net building and testing rn. After several hours of work
the 386bsd closed connection and I was unable to reconnect. I tried to get
in from the other side of the slip connection through our LAN and was unable to
reach the 386bsd host from there as well. It had apparently crashed. Later
in the morning I had someone hit the reset button on it (I was home). It
didn't come back up.
When I got to the office it had booted to the point where the next
message should have been that it was mounting root on /dev/as0a. There it sat.
I got my Fixit floppy (patchkit 0.2.1 level, patches 1-83) and booted with it,
did an fsck against all the filesystems on both disks (no major uglies
detected) and then re-ran fsck again - all was clean. I tried to boot again
and still the same result as before. Next I tried verifying and rewriting the
disklabel and boot blocks. That went smoothly but didn't help either.
Finally I tried recopying my 386bsd.alt kernel (Patch level 1-83) over my
normal kernel 386bsd (Patch level 1-110). Still no luck.
I have no trouble mounting and using any of the filesystems on either
of the two disks as long as I boot from floppy. I checked and verified that
both init and rc are present on in their proper places on the root filesystem.
Is there anything else I can try short of reinstalling? Does it sound like
I have a hardware problem? I was thinking about swapping the Quantum
80MB (as0) for another and reinstalling to that first. I'd actually love
thoug for someone to tell me that all I have to do is use some cryptic piece
of unix wizardry and obscure utility to fix some obtusely confused area
of my disk that even fsck doesn't care about...
On another issue, When booting with the the patch 1-100 kernel (Julian
SCSI driver) I sometimes get the following:
aha0 reading board settings, trap type 19 code = 0 eip = fe000744 cs = 8 \
eflags
= 246 cr2 0 cpl ffffffff
panic: trap
hit reset please
Hitting reset usually gets me by this. It seems to happen both on power-ion
and warm boots. Even so, when it booted it was usually fine after that. This
problem doesn't seem to be related to the current crisis (as far as I can
tell). Lastly, when the system does boot normally, Julians SCSI driver says
something about something needing to be 200ns instead 250ns. I could provide
the exact text if needed. Does anyone have any words of advice?
Thanks VERY much in advance for your thoughts!
-John
jackson@a1.mec.mass.edu or vikki@athena.mit.edu
(and, when it's stable, toad@lobster.sid.mcet.edu...)