*BSD News Article 35027


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!msuinfo!agate!howland.reston.ans.net!torn!news.unb.ca!jupiter.sun.csd.unb.ca!b6ps
From: b6ps@jupiter.sun.csd.unb.ca (Peter Howlett)
Newsgroups: comp.os.386bsd.bugs
Subject: Re: FreeBSD, fd0d: hard error
Date: 27 Aug 1994 14:17:34 GMT
Organization: University of New Brunswick, Fredericton, NB, Canada
Lines: 40
Message-ID: <33nhtu$3fa@sol.sun.csd.unb.ca>
References: <33j9hp$41l@sol.sun.csd.unb.ca> <1994Aug26.044412.20941@news.csuohio.edu>
NNTP-Posting-Host: jupiter.unb.ca
X-Newsreader: TIN [version 1.2 PL2]

Steve Ratliff (stever@csuohio.edu) wrote:
: Peter Howlett (b6ps@jupiter.sun.csd.unb.ca) wrote:
: : Howdy,

: : Has anyone ever seen this?

: : fd0d: hard error reading fsbn 62 of 64-71
: : (ST0 44<abnrml,top_head> ST1 20<bad_crc> ST2<bad_crc> cyl 1 hd 1 sec 15

: 	This pretty much looks like a bad spot on the floppy to me.  At
: a guess ;)  I would say at cyl 1 head 1 sector 15.

: Try making a new cpio floppy on a 100% error free floppy.  I've come to
: the hard learned conclusion that 3M pre-formated guarantee'd floppies

: Steve

I have tried this again. I tried it with some nice new sony's, a different
box of my generic's and with the old floppies from before. Each time I
get the same thing in different places. Sometimes I get the floppy errors
when the disklabel is being written, sometimes it makes it through most
of the cpio'ing of the filesyst disk, but it always appears somewhere.
If is does make it through the process only printing the messages and
not hanging, I boot the kernel floppy again and copy the kernel to
the disk. We still get problems after that because it seems as though
what got copied to the disk is not complete, and it doesnt boot. (The
geometry is right because it is printed in little letters right on the
drive, a Quantum LPS540S.

I did actually try to boot linux, it does work with the 1542 driver
and loaded without any problems (on some of the same floppies). 
Perhaps, as suggested by someone who wrote me with e-mail, the SCSI
is hogging bus bandwidth to long and blowing the floppy DMA? He mentioned
that this was an Adaptec problem though, could it happen with a 
Buslogic EISA SCSI? (BT747c, brand new) I personnally am the wrong one
to ask, but I mention this in case anyone is interested. 

-------------------------------------------------------------------
     From:  Peter Howlett               University of New Brunswick  
     Inet:  b6ps@unb.ca			Fredericton, N.B. Canada