PCBUILD Archives

Personal Computer Hardware discussion List

PCBUILD@LISTSERV.ICORS.ORG

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Tom Turak <[log in to unmask]>
Reply To:
PCBUILD - Personal Computer Hardware discussion List <[log in to unmask]>
Date:
Fri, 19 Feb 1999 16:13:23 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (92 lines)
Most of this rambling message is further proof that I don't have enough
do at work.  Anyway, I couldn't find the debug command so I
am glad to see David Gillette suggest it.  After reading the
other suggestions, I might add, if you haven't run scandisk
"thorough test", don't, it probably won't cure anything and it may
mark the drive up with bad sectors.
As it stands it sounds as if the drive is unusable, so there is nothing
lost by trying something.  I will assume this is the original drive,
and not a replacement IDE drive someone else installed.  IDE
drives like a 40 meg Seagate ST-351 A/X require special software
to low level format, although Seagate FAQ says a CMOS routine with
an AMI BIOS date greater than 1991, will work on their IDE drives,
but only an AMI. I would be against using this on an ATA-2 or UDMA
drive, regardless. At a 40 meg size, your drive is definitely a plain
vanilla IDE or an MFM drive. I also have a repair textbook here that
says you can't damage an IDE drive by low level formatting, but
you can destroy some embedded optimizations the factory
installs on the drive, hurting the disk's performance.
It also says you can't hurt a drive that using "zoned bit recording"
because it uses a translation scheme that "must" be protected,
consequently the factory installs stuff that protects it from tampering.
The text also says it is mostly pointless to low level format
such a drive because essentially all it accomplishes is a total
erasure, no error correction or realignment or refresh, whatever.
Seagate FAQ agrees that low level formatting their IDE drives
doesn't serve any useful purpose.  The FAQ specifically says to answer
"No" if the AMI Bios routine requests to update the bad block table
for IDE drives, which I found interesting because the same routine
adjusted the total capacity of my MFM drives after it finished, which I
assume was due to more bad blocks found and marked.
I had a problem similar to yours with Seagate ST-251 40 meg MFM drives.
You can get by just by formatting the drive for a while.  Your
problem now is the error appears in the directory listing, rather
than in a data portion of the disk. After reformatting, I dealt with the
inevitable error messages by changing the name of the effected file
to BAD001, the next error to BAD002, effectively blocking out bad sectors.
You can't do that if the error appears in the directory sectors, fat sectors, etc.
So you may be satisfied with just a regular formatting.
I finally fixed the problem permanently, ( I still use the drives as
DOS partitions in old SCSI 486's) by moving the drives into a 486
system that had the low level format routine in its CMOS BIOS menu.
A 386 will work too, in most cases.  Your 286 may even have the routine
if it is a clone built toward the end of the 1980's, not a compaq or IBM.
If it doesn't, You will need to take the controller card out of the 286 and
install it in the newer computer too.  You will probably need access to your
286 CMOS BIOS because if everything else works, the 286 may not recognize
the drive without you going through setup again. If you don't know how to get
into CMOS setup for the 286 I would stop right now.  Definitely write down
what the drive configuration is in the 286 setup BEFORE removing the drive.
Once you get the drive into the 486, (I would disconnect the 486's own drive
as a precaution, and make sure you record the 486 PC hard disk parameters
before you open its case too.) turn it on, and hit DEL to enter setup.
Setup the 486 drive C: parameters to match the 286 hard disk, and
choose Format Hard Disk, and see what happens.  Move the drive back into
the 286,  boot from a dos floppy. set the heads, cylinders, and sectors back
up if necessary in setup, run fdisk and set a primary dos partition, re-boot
with the floppy again and  Format C: /s.  This is an awful lot of trouble compared to
using a debug routine, but I found it worked, I liked the friendly user interface,
and based on the fact that my two are still running after 10 and 11 years, respectively,
it was worth it.  And of course the cost was free.  They should be in a landfill.
My only complaint with the SCSI 486's (they don't need screens or keyboards
for what they do) is that they pull a lot of watts compared to a modern system,
so they will most likely go to the landfill still in working order.

Tom Turak

On Wednesday, February 17, 1999 7:35 AM Dave Thuillier  wrote:

I can boot to the C: drive and see the directory (dir/w) without a
problem. I can run other programs that are on the hard drive and
in other directories. The previous owner had lotus, windows 3.1,
a couple of games, and some misc other software installed.

However, if I change to the "windows" directory (win 3.1) and then
try to view the directory the following errror message pops up.
"general failure reading drive C: abort, retry, fail?"

This error message pops up every time in the "WINDOWS" directory
only. Any ideas?

I have the disks for dos 6 and win 3.1. I am just
not sure wether they are still good. Do I need to reformat first?
If so how do I do it? I have seen warnings about formatting a hard
drive on this list in the past few weeks and don't want to kill
a potentially useful computer.

Dave Thuillier

                Curious about the people moderating your
                   messages? Visit our staff web site:
                     http://nospin.com/pc/staff.html

ATOM RSS1 RSS2