18 Jan
2005
18 Jan
'05
6:35 a.m.
Hi
DOS has an irritating feature when asked to format
a disk. If track 0 is partially readable it will
fail to format, even if you use the /u option.
I've been moving data from one machine to another
and the source machine has a flaky drive. It sometimes
trashes track 0.
I've found that the only way to get around this problem
is to wipe the disk with a strong magnet.
Why can't it just try to format first and then check
track 0. Why must it fail to format because track 0
is partially readable when I specifically asked it
to unconditionally format?
Dwight
18 Jan
18 Jan
6:47 a.m.
From: "Dwight K. Elvey" <dwight.elvey(a)amd.com>
Hi
DOS has an irritating feature when asked to format
a disk. If track 0 is partially readable it will
fail to format, even if you use the /u option.
I've been moving data from one machine to another
and the source machine has a flaky drive. It sometimes
trashes track 0.
I've found that the only way to get around this problem
is to wipe the disk with a strong magnet.
Why can't it just try to format first and then check
track 0. Why must it fail to format because track 0
is partially readable when I specifically asked it
to unconditionally format?
Dwight
Try picking up the format program from freedos/opendos and see if it does
the same. If needed it may be able to be modified.
Randy
7:34 a.m.
On Tue, 18 Jan 2005, Dwight K. Elvey wrote:
DOS has an irritating feature when asked to format
a disk. If track 0 is partially readable it will
fail to format, even if you use the /u option.
I've been moving data from one machine to another
and the source machine has a flaky drive. It sometimes
trashes track 0.
I've found that the only way to get around this problem
is to wipe the disk with a strong magnet.
Why can't it just try to format first and then check
track 0. Why must it fail to format because track 0
is partially readable when I specifically asked it
to unconditionally format?
I know. It's another one of those stupid things that MS did with DOS that
makes absolutely no sense. Just format the damn disk and report any
errors for the USER to decide.
I once came across some program on the WWW that supposedly rehabilitates
such disks (most likely by doing an LLF on track 0) but unfortunately I
never bookmarked it.
--
Sellam Ismail Vintage Computer Festival
------------------------------------------------------------------------------
International Man of Intrigue and Danger http://www.vintage.org
[ Old computing resources for business || Buy/Sell/Trade Vintage Computers ]
[ and academia at www.VintageTech.com || at http://marketplace.vintage.org ]
10:20 a.m.
On Tue, 18 Jan 2005, Dwight K. Elvey wrote:
track 0. Why must it fail to format because track 0
is partially readable when I specifically asked it
to unconditionally format?
Because MICROS~1 is completely unclear on the concept of "unconditional".
In their more recent operating systems, they are also completely unclear
on the concept of "floppy boot". If one of the system files ON THE HARD
DISK is bad, it CAN NOT be booted. What else can I boot an NT machine
with that can read and write an NTFS partition?
10:46 a.m.
On Tue, 18 Jan 2005 13:20:41 -0800 (PST), Fred Cisin <cisin(a)xenosoft.com> wrote:
On Tue, 18 Jan 2005, Dwight K. Elvey wrote:
Checkout Knoppix it can do loads of stuff with ntfs partitions. Virus
checking, repairing, password changing.
Dan
track 0. Why must it fail to format because track
0
is partially readable when I specifically asked it
to unconditionally format?
Because MICROS~1 is completely unclear on the concept of "unconditional".
In their more recent operating systems, they are also completely unclear
on the concept of "floppy boot". If one of the system files ON THE HARD
DISK is bad, it CAN NOT be booted. What else can I boot an NT machine
with that can read and write an NTFS partition?
10:50 a.m.
At 03:20 PM 1/18/2005, Fred Cisin wrote:
In their more recent operating systems, they are also
completely unclear
on the concept of "floppy boot". If one of the system files ON THE HARD
DISK is bad, it CAN NOT be booted. What else can I boot an NT machine
with that can read and write an NTFS partition?
Any one of the Linux-based boot CDs that handle NTFS? Or the
Sysinternals tools that handle NTFS? Or a second machine that
can still read the hard disk, if you transplant it?
- John
4:10 p.m.
>In their more recent operating systems, they are
also completely unclear
>on the concept of "floppy boot". If one of the system files ON THE HARD
>DISK is bad, it CAN NOT be booted. What else can I boot an NT machine
>with that can read and write an NTFS partition?
On Tue, 18 Jan 2005, Dan Williams wrote:
Checkout Knoppix it can do loads of stuff with ntfs
partitions. Virus
checking, repairing, password changing.
Thank you
An excellent possibility. I just ordered the disk, but might try
downloading it tomorrow when I have access to a faster connection.
On Tue, 18 Jan 2005, John Foust wrote:
Any one of the Linux-based boot CDs that handle NTFS?
I just ordered Knoppix
Or the Sysinternals tools that handle NTFS?
Unless I missed something,...
They have lots of neat stuff that runs under NT, and
they have two different programs that can READ NTFS from a 98 boot.
One of them is a nice amateur attempt at an XTREE like interface, but READ
ONLY. And it won't read SOME NTFS partitions (8G limit?)
The other is an excellent working device driver, that deliberately won't
write, to permit them to SELL that version.
To fix the machine so that it can boot, I need to copy a good copy of
NTVDM.EXE to it.
Or a second machine that
can still read the hard disk, if you transplant it?
That's what I'll be doing next.
--
Grumpy Ol' Fred cisin(a)xenosoft.com
19 Jan
19 Jan
5:22 a.m.
>>>> "John" == John Foust
<jfoust(a)threedee.com> writes:
John> At 03:20 PM 1/18/2005, Fred Cisin wrote:
> In their more recent operating systems, they are
also completely
> unclear on the concept of "floppy boot". If one of the system
> files ON THE HARD DISK is bad, it CAN NOT be booted. What else
> can I boot an NT machine with that can read and write an NTFS
> partition?
John> Any one of the Linux-based boot CDs that handle NTFS? Or the
John> Sysinternals tools that handle NTFS? Or a second machine that
John> can still read the hard disk, if you transplant it?
Linux NTFS has an "experimental" read/write capability for NTFS, but
that only works with NTFS V4 -- with WinNT. It doesn't handle the
Win2k or later flavors of NTFS. Or at least, it didn't last I looked
(which I think was Fedora 1).
paul
18 Jan
18 Jan
12:25 p.m.
for media portability the track 0 was the directory on some
IBM formats, and probably the dos program is trying to bridge
the problem by forcing the whole track to have readable sectors
on it, so some other non dos programs won't blow.
I had experience doing a 8" driver at one time and was surprised
to find that the standard format for ibm was to have at least the first
sector present, but the rest could be "deleted" sectors, the lot of
which
could be replaced with other data sectors, and a string of deleted
ones again to add data.
Also track 0 frequently tells software that the format is for the entire
disk, and may have been assumed to be in a standard density, with
tracks 1 thru n in a different density, determined from track 0.
I think on 5 1/4" diskette formats IBM did look at a byte in the
first sector to tell 320 vs 360 or 160 vs. 180 k format. Later
bios' assumed a 360k format if the byte was not FF or FE, IIRC.
Some non dos os's were smart when designing their disk formats
to be used inthe IBM pc and kept the media byte, and some,
(like Pick for the PC) were dumb and had diskettes which looked
totally unformatted to the bios and to DOS when set up for
their use.
Jim
"Dwight K. Elvey" wrote:
Hi
DOS has an irritating feature when asked to format
a disk. If track 0 is partially readable it will
fail to format, even if you use the /u option.
I've been moving data from one machine to another
and the source machine has a flaky drive. It sometimes
trashes track 0.
I've found that the only way to get around this problem
is to wipe the disk with a strong magnet.
Why can't it just try to format first and then check
track 0. Why must it fail to format because track 0
is partially readable when I specifically asked it
to unconditionally format?
Dwight
10:28 p.m.
On Tue, 18 Jan 2005 09:35:01 -0800 (PST), Dwight K. Elvey
<dwight.elvey(a)amd.com> wrote:
Hi
DOS has an irritating feature when asked to format
a disk. If track 0 is partially readable it will
fail to format, even if you use the /u option.
I've been moving data from one machine to another
and the source machine has a flaky drive. It sometimes
trashes track 0.
...
Any OS / file system has problems if track 0 is really bad, especially if
sector 0 is bad.
I have been able to "fix" old IBM 10MB drives by adjusting the physical
end stop by a fraction of a trackwidth.
On these drives, there is a small bumper on the outside that the external
part of the actuator bumps against when you send the drive a command to
recalibrate or locate to cylinder zero. Adding a very thin sliver of metal
will move the calibration point by a fraction of a track. If you try this,
you may be able to find a spot where you can successfully format the first
cylinder.
--
Bj?rn
19 Jan
19 Jan
5:39 a.m.
>>>> "Bj??rn" == Bj??rn Vermo
<bv(a)norbionics.com> writes:
Bj??rn> On Tue, 18 Jan 2005 09:35:01 -0800 (PST), Dwight K. Elvey
Bj??rn> <dwight.elvey(a)amd.com> wrote:
> Hi DOS has an irritating feature when asked to
format a disk. If
> track 0 is partially readable it will fail to format, even if you
> use the /u option. I've been moving data from one machine to
> another and the source machine has a flaky drive. It sometimes
> trashes track 0.
Bj??rn> ...
Bj??rn> Any OS / file system has problems if track 0 is really bad,
Bj??rn> especially if sector 0 is bad.
That's definitely not true. For example, PDP11s use sector 0 for the
boot sector, bot not for anything else. A disk that doesn't need to
be booted will work perfectly well with an unreadable sector 0.
Depending on the OS, there may be some other critical sector (sector 1
on RSTS, for example). More sophisticated file structures have a
search rule, where the "superblock" can be found in one of several
places that are checked in order.
paul
7158
days inactive
7159
days old
10 comments
9 participants
participants (9)
-
bv＠norbionics.com -
cisin＠xenosoft.com -
dwight.elvey＠amd.com -
jfoust＠threedee.com -
jwstephens＠msm.umr.edu -
pkoning＠equallogic.com -
randy＠s100-manuals.com -
vcf＠siconic.com -
williams.dan＠gmail.com