How do I low-level-format an IDE Drive?
Frank Overstreet
install SCO on. After installing SCO, the PC won't boot. The error
messages cause me to believe root or some portion of root is located
beyond the 1024th sector. I tried to re-compute a smaller size for root
but without success. Now I want to low level format the drive and am
wondering if the Western Digital utility wddiag.exe is what I need. When
the readme describes writing all 00's is that the same as a
low-level-format. If not please help.
Regards,
Frank
Jeff Liebermann
My car won't start. How do I rebuild the engine?
If you want a big hammer solution, have fun:
http://www.wdc.com/service/ftp/drives.html#wd_diag
http://www.wdc.com/service/ftp/drives.html#wd_diagUG
Don't say I didn't warn you. WDDIAG will certainly clean off the drive.
Be sure to let the drive warm up to operating temperature (approx 15 min)
before running WDDIAG or you will have thermal drift problems.
However, what you apparently want is just a way to clean off the drive and
start over. Fire up the MSDOS 6.22 FDISK program and "remove a non-dos
partition". If you also want to vaporize the boot loader, run:
FDISK /MBR (master boot record).
This will work for everything exept NT4 which requires the NTREMOVE program
from the NT Server Resource Kit (I think it's on the update 2 cdrom).
Alternately, you can use "dd" or a DEBUG script to overscribble the boot
record.
--
Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
(831)421-6491 pgr (831)426-1240 fax (831)336-2558 home
http://www.cruzio.com/~jeffl WB6SSY
je...@comix.santa-cruz.ca.us je...@cruzio.com
Bob Bailin
Normal mode. If you use LBA mode, the drive is *always* configured to have
less then 1024 logical cylinders, so the boot limitation problem you describe
just won't happen.
Be aware that when you change the disk configuration from Normal to LBA, all
data on the drive will be effectively lost.
You cannot "low-level format" modern EIDE drives, however, the wddiag.exe
program will scan (non-destructively or destructively) the drive for errors
and attempt to re-assign any bad blocks it may find. You must run wddiag
from a DOS or Win9x bootable floppy, not from the hard disk you are testing.
Frank Overstreet <fover...@mindspring.com> writes:
>
> I have a Western Digital Caviar 2.1G IDE drive which I'm trying to
> install SCO on. After installing SCO, the PC won't boot. The error
> messages cause me to believe root or some portion of root is located
> beyond the 1024th sector. I tried to re-compute a smaller size for root
> but without success. Now I want to low level format the drive and am
> wondering if the Western Digital utility wddiag.exe is what I need. When
> the readme describes writing all 00's is that the same as a
> low-level-format. If not please help.
>
> Regards,
> Frank
>
--
Bob Bailin
72027...@compuserve.com
Tony Earnshaw
> I have a Western Digital Caviar 2.1G IDE drive which I'm trying to
> install SCO on. After installing SCO, the PC won't boot. The error
> messages cause me to believe root or some portion of root is located
> beyond the 1024th sector. I tried to re-compute a smaller size for root
> but without success. Now I want to low level format the drive and am
> wondering if the Western Digital utility wddiag.exe is what I need. When
> the readme describes writing all 00's is that the same as a
> low-level-format. If not please help.
If you even attempt to low-level format a modern IDE disk, you'll ruin
it. This has long been the case. Writing all 00s is not the same as a
low-level format, it's what it says.
Low level formatting was possible, after manufacture, with (now) almost
prehistoric drives to remagnetize surfaces (thereby removing corruption
and thus sometimes repairing some bad spots on the surfaces). These
drives did not carry translation tables, as modern drives do.
http://www.western-digital.com/service/diagnostics.html reports wddiag
as simply being a diagnostic utility that can erase all data on the
drive to remove any corruption - a normal format does not do this.
For the rest, you do not make any mention of what OS version you're
using, other hardware, disk configuration or anything else. No one can
help you if you don't.
Tony
--
************* THE NEW DIMENSION IN DISTRIBUTION ***********
ilion Faculty B.V.
Tony Earnshaw email: to...@ilion.nl
Randstad 21-57
1314 BH Almere-Stad tel: +31 (0) 36 548 50 10
The Netherlands fax: +31 (0) 36 534 05 34
***************** http://www.ilion.nl *********************
Bill Vermillion
<to...@ilion.nl> wrote:
>Frank Overstreet wrote:
>> ... Now I want to low level format the drive and am wondering
>> if the Western Digital utility wddiag.exe is what I need. When
>> the readme describes writing all 00's is that the same as a
>> low-level-format. If not please help.
>If you even attempt to low-level format a modern IDE disk, you'll ruin
>it. This has long been the case. Writing all 00s is not the same as a
>low-level format, it's what it says.
>Low level formatting was possible, after manufacture, with (now) almost
>prehistoric drives to remagnetize surfaces (thereby removing corruption
>and thus sometimes repairing some bad spots on the surfaces). These
>drives did not carry translation tables, as modern drives do.
The old wives tale of 'remagnetizing' the drives is just a myth.
Magnetic media is quite stable - it's the environment that does
them in re-acting with binders in coated media. Media for the most
part is plated/sputtered today, so the only problem is the decaying
of the particles. It just doesn't happen - at least in a computers
lifetime. This excludes catastophic events of course, and high
heat levels - above 150F you are going to have problems.
One of the ways the myth seemed to get started was on the old MFM
drives of the ST-506 heritage. These were all 'stepper' drives.
eg - a motor turned x degrees and ratchets the head across the
drive surface. (In the floppy arena it was typically to have
to re-aline a 5.25" disk every 6 months when used in heavy duty
service. I did that but was pushing them 24x7x365. The first
drives would last about a year, and when the 1/2 heights came out
you could expect 4 years approximately - MTBF was about 20,000
hours for those).
The mechanism would wear over time and when the drive was issued
commands to pulse/step the drive to the track, after a time it
the head would not be positioned exactly in the center of the track
set by the original format, and a reformat would then bring back
the performance as first seen as the platter to stepper were now in
sync with the worn portions.
To try to improve performance embedded servos were being used.
This was a servo burst in between sectors. Doing a real low-level
format meant the drive had to go back to the factor for a new
format and servo. It was expensive. Typically the servo looked
like a 'wedge' if you viewed it magnetically as the outer tracks
had the bits spread further apart.
Then came the dedicated servo drive - with the bottom platter
being used only for servo. This is why you'd see drives
with and even number of platters, but one less than the total
for data.
These are the drives that perform the thermal recal because as the
enivornment changes the metals contract and expand and the bottom
head is controlling the position of all other heads on the stalk.
Current technolgy is embeded servo again - but there's no way a
user can screw these up - as the old drives were controlled by
cards external to the drive, and the new ones are integral to the
units.
This eleminated thermal recallibration, ZBR (zone bit recording)
gives a different number of sectors available on different track
groups.
Low-level reformating really needs never to be done. Worst case -
to get rid of some pesky droppings by some ill-behaved program, or
programming concept, would be the destructive verify in the
controller.
But 'reformatting to refresh the format' is something left over
from DOS circa 1985.
--
Bill Vermillion bv @ wjv.com
Tony Lawrence
and I hate to see them disappear into the ether.
Anytime you'd like to put some of these into article form, I'd be happy
to publish them..
--
Tony Lawrence (to...@aplawrence.com)
SCO ACE
SCO articles, help, book reviews: http://www.aplawrence.com
Bill Vermillion
<to...@aplawrence.com> wrote:
>You know, Bill, you do a really excellent job with some of your
>replies, and I hate to see them disappear into the ether.
>Anytime you'd like to put some of these into article form, I'd be
>happy to publish them..
Well thank you for the compliment. As you may have noticed I get
prolific from time to time :-) . It goes back to the days when I
was paid to talk about 20 minutes/hour for about 6 hours per day, 7
days a week.
I also find the net/email the most convenient form of
communications since a computer group I was in back in 1980 used
email on "The Source" to communicate. Back then there were about
3000 users on the Source, and when they screwed up their email by
trying to add a second Prime and took response to sometimes 5
minutes !! we all moved to MicroNet - and there were about 3000
people there too. They were since renamed Compuserve.
Everything I put on the net is 'fair game' - as you notice no
copyrights, but if someone reprints it I'd at least like some
credit. I've seen my name in print often enough to realize
that 'mis-quoting' is a way of life it seems. Billboard never
could seem to get things correct, so I can truly sympathize
with those who say "I've been misquoted". Of course I worked in
broadcast and it seemed to be that at least one person was
recording what you did at any time so you could never deny that you
said it.
I treat my articles just as you would a newpaper article, here
today, something else in it's place tomorrow.
I don't have time to worry about what I've said in the past.
The only time someone persuaded me to re-edit and turn it into
an artilce was on the old rec.audio.pro group, having to
do with magnetic tape alignment. I got published in their
newsletter along with such notables as Carver.
If you feel like sticking these into your site, or anyone else
feels the same, feel free. You'd have to convince me of the worth
of re-writing for net distribution. I'd consider it for other
media.
Bill