Reinitializing the bad sector count of a HD?
Bill in Co
to the reserve pool), assuming they were just soft errors, due to a power
outage (not physically damaged sectors on the disk)? Or is this
permanently locked in by the drive's firmware?
Cheng Heng
Format means to prepare a storage medium, usually a hard drive,
for reading and writing. When you format hard drive, the operating
system erases all bookkeeping information on the disk, tests the
disk to make sure all sectors are reliable, marks bad sectors, and
creates internal address tables that it later uses to locate
information.
hth
Hot-Text
But run Error Checking Status
Chick Thorough
Chick Options
Check Do not perform write-testing
chick OK
Check Automatically fix errors.
Chick Start
if there "bad" Cluster in sectors it will move the data to a good Cluster
and mark "bad" Cluster so no data will be write to it.
Then Defragment Now
to move the data Cluster back to the same sectors and open up the Free
space!
And you reformatting a hard drive..... run Error Checking
Status.............Defragment Now........
Best to be Safe then Sorrow for not doing so!
"Bill in Co" <surly_cu...@earthlink.net> wrote in message
news:jM-dnfxT9ZhELkjR...@earthlink.com...
Dennis
I have salvaged at least a dozen hard drives with it. It doesn't move
the info and mark the sector bad, it repairs bad sectors.
http://www.dposoft.net/
John John - MVP
Formatting will not reset the bad sectors replaced by the firmware, it
will only reset the file system's bad cluster list, the $BadClus file.
Bad sectors replaced by the spare pool are invisible to the operating
system, unless you are using SCSI disks the operating system doesn't
know about sector sparing, the firmware silently hides and remaps these
sectors.
John
Bill in Co
On this one hard disk I was using I was getting a HD Sentinel Disk Health of
97% due to 2 "bad sectors" (which were created during a power outage). But
those sectors aren't really hardware defects, so I'd like to get the Disk
Health (and reported 2 bad sector count) cleared up.
So I'm confused - will reformatting the disk be able to do that, John?
Bill in Co
repairs bad sectors for hardware defects, but this isn't a hardware defect
in my case.
Jason
Jason
hard disks but there is a program designed to format floppy disk and
test each sector to the extent that the built in Microsoft format
facility can't. It's probably like the DOS 6.22 and earlier
format.exe/format.com but runs under windows. I had disks that were
giving unformatted error or another error that stopped the disk being
read and this program got them working. I haven't had issues with hard
disks other than what the /f switch does for a long time.
John John - MVP
How can you be certain that they are not hardware defects? Isn't it
possible that the sectors may have been physically damaged when the
power failure occurred? What does Chkdsk report? If chkdsk is
continually reporting bad sectors then the disk is failing! Did you run
the disk diagnostic utility from the disk manufacturer? What does it
report?
I take it this is a data disk? If it is only a software (file system)
error then formatting will reset it, if it was marked off as a bad
sectors by the firmware formatting will not reset it as the marked off
sectors are invisible to the operating system or to the formatting
utility. The answer to your question will be revealed when you format
the disk, let us know the outcome.
John
John John - MVP
program.
Jim
Quote>
How it works
Almost 60% of all hard drives damaged with bad sectors have an
incorrectly magnetized disk surface. We have developed an algorithm
which is used to repair damaged disk surfaces. This technology is
hardware independent, it supports many types of hard drives and
repairs damage that even low-level disk formatting cannot repair. As a
result, previously unreadable information will be restored. Because of
the way the repair is made, the existing information on the disk drive
will not be affected!
Unquote>
John John - MVP
I'm still somewhat skeptical. As a data recovery tool it may do a very
good job of (attempting) to recover data from bad sectors. As a tool to
"fix" a hard drive with bad sectors marked off by the firmware I
wouldn't really rely on it. Firmware will mark a sector bad if it can't
read the sector after x retries, often the sector can be read after more
retries but the firmware does not see the sector as trustworthy because
it needs too many retries to read the sector so it prudently marks it as
bad. I really think that all that the mentioned program does is try to
read the bad sector for a much greater length of time than the firmware,
it probably just keeps on 'hammering' away at the sector for a a
considerable length of time before it gives up. Personally I wouldn't
trust my data to a drive 'fixed' with this software.
John
Paul
http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=203931&NewLang=en
"What does "low level formatting" a SATA or ATA (IDE) drive mean?
Actually the term "low level" is a bit of a misnomer. The low-level process
first used years ago in MFM hard drives bears little resemblance to what we
now call a "low-level format" for today's SATA and ATA (IDE) drives. The only
safe method of initializing all the data on a Seagate device is the Zero Fill
erase option in SeaTools for DOS. This is a simple process of writing all zeros
(0's) to the entire hard disk drive.
Why would I want to Zero Fill my drive?
The most common reasons to Zero Fill a SATA or ATA (IDE) hard drive are:
* to remove a virus that cannot be removed without destroying the boot sector.
* to change from one operating system to another and wish to remove everything
from the drive.
* to erase confidential information for privacy reasons.
* to scan for bad sectors that can be detected and replaced with good spare
sectors when writing to the sectors. <----- not what Bill wants...
By design, modern disc drives maintain spare sectors for reallocation purposes.
Usually, sectors become difficult to read long before they become impossible
to read. In this situation the actual data bytes in the sector are preserved and
transferred to the new spare during a sector reallocation. Similarly, when a disc
drive writes data (like a zero fill erase procedure) and encounters a problem, the
drive firmware retires the problem sector and activates a replacement before giving
successful write status."
So what that KB article tells you (and it isn't 100% accurate), is
that
- there is no longer a low level format command. Modern drives use embedded
servo. Servo --- DATA --- Servo --- DATA --- Servo is the track
pattern. An end user can only write to the DATA section. The pattern
of servo marks is applied at the factory. They're called "servo wedges",
because of the pie shaped pattern they make when viewing the platter from
above.
- the zeroing option is via normal write operations. A write operation, applied
to a sector on the pending list, potentially spares out the sector (and adds
to the reallocations on the disk). So zeroing will only make the situation worse,
with respect to "improving the statistics for the drive". It can't reduce
the number of already occurred reallocations.
*******
SCSI drives have a different history than IDE. This modern SCSI drive I could find
for sale at Newegg yesterday, still has P-list and G-list mentioned in the
documentation. It implies, that if you have the right maintenance utility, it
would be possible to reset reallocations. When I used SCSI drives on my
home systems, finding a decent utility at a reasonable price, was always
an issue, and I had to buy drives based on the drive showing up in the
utility as being supported. Not a good situation at the best of times.
( See section 8.1 on PDF page 41, for defect list management on SCSI. )
http://www.seagate.com/staticfiles/support/disc/manuals/enterprise/cheetah/15K.5/SCSI/100384776j.pdf
So if Bill had a SCSI drive, it could be fixed. (I think at work, I reset
the G-list on at least one drive, so I've done it. I don't know what utility
I used at the time, that's how long ago it was.) With an IDE drive, there
are fewer chances of blowing away any reallocations done while the disk
was running.
*******
IDE drives do have various "private" interfaces. This article describes a
recipe for recovering a particular kind of fault on a Seagate 7200.11 drive.
It uses a TTL level serial port on the drive, to deliver "cryptic commands"
to the drive. I don't know where the document would exist, with a list of
what commands are available. Suffice it to say, the cryptic command set
is not the same as ATA, and has more to do with internal drive operation.
I would have expected any Seagate documentation describing this interface,
to be covered by an NDA.
http://sites.google.com/site/seagatefix/
http://www.overclock.net/hard-drives-storage/457286-seagate-bricked-firmware-drive-fix-pics.html
So if anything is going to work, that's an example of how you'd do it.
If Franc was around, he'd probably have some better answers.
http://forums.seagate.com/t5/Barracuda-XT-Barracuda-and/Seagate-Barracuda-dead/m-p/44864
Paul
Bill in Co
damage).
But according to its website, it allegedly has the "ability to repair
physical bad sectors (magnetic errors) on a hard disk surface".
So the confusion here is over the use of the term "physical". :-)
In retrospect, I think you're right, and the use of the term "physical" is a
bit misleading. I'm assuming the only thing it can do is rewrite data to
the disk in the so called bad sector areas, and somehow hope that that fixes
it. Clearly any areas that had true physical damage (scratches, holes, or
whatever) wouldn't be helped.
Bill in Co
As far as chkdsk was concerned, everything is/was fine.
But HD Sentinel reported 2 "bad" sectors, and SMART reported 2 reallocated
sectors in its reallocated sector count data value. HD Sentinel
consistently gave this disk a 97% Disk Health status (I've had NO problems
with this SATA HD, which was my main system HD, and I only wanted to clear
up the "alleged" damage created by that one power failure).
Well, given what you, Paul, and others have said, it doesn't sound like
reformatting would do anything for this. IOW, that reallocated/bad sector
status been locked in by the firmware for all eternity, and I'd be chasing
windmills.
Last night I went ahead and cloned this drive (my system SATA drive) over to
a brand new and identical SATA HD on this Dell Desktop using Acronis True
Image, and this new disk has NO bad sectors, naturally. :-)
This one works great, except that I lost the ability to ever use Ctrl-F11 to
restore back to the factory setup partition, if I ever wanted to (not
likely).
In fact, pressing Ctrl-F11 at bootup now makes the system unbootable, so I
then have to run DSRFIX on a floppy to fix that, anytime I dare try it).
I found this all a bit surprising, since I was under the impression that
making a HD clone (I used Acronis True Image) made an identical copy of
*everything* with data in it - guess not.
I could have chosen to use the (much more time intensive) sector-by-sector
cloning process, but decided against it, figuring that I didn't need to do
that. And I don't, so long as I give up the Dell Restore to factory setup
fallback capability.
Still, it was a bit annoying that ATI missed copying something in its
cloning, but I'm not sure what it is. I'm thinking it's something related
to the partition tables and MBR, but thats still a bit unclear to me.
Paul
Dell has all sorts of tricks, and this site has good info on dealing with it.
http://www.goodells.net/dellrestore/mediadirect.htm
The first thing I'd want to understand, is whether your drive has an HPA
or Host Protected Area defined. If so, then you're in for a real challenge.
Paul
Bill in Co
Thanks Paul, and I'm well aware of Goodell's site. :-) That's where I got
the DSRFIX (Dell System Restore) fix utility from. I don't know anything
about the Host Protected Area though - that's the first time I've heard of
that term. I guess I'd better look that up.
The interesting thing is, everything is fine here unless I try using that
Ctrl-F11 bootup option (to allow access to the Dell system restore option if
one wanted it), and if I use Ctrl-F11, it ends up locking up the system, so
I then need to reboot and run DSRFIX on a floppy so that I can ever boot up
again. The active windows partition here is #2, I think, since at bootup,
Dell's hidden partition initiates everything from startup to go through it
first before it passes control over to windows.
But under no circumstances am I ever able to successfully run the Ctrl-F11
restore option anymore (it fails and locks up the system at the BIOS level,
and then I have to "fix it" with DSRFIX (Goodell's Dell System Restore
utility fixer), which at least allows me to reboot into windows again!).
Clearly SOME code was NOT cloned over from the original drive using Acronis
True Image's cloning option. I'll probably go back and restudy Goodell's
site again, but I think this is irreparable without editing some disk
sectors manually, and I don't want to (or know enough to) "go there". At
least I knew enough to use DSRFIX to be able to reboot again after trying
Ctrl-F11 which messed up the partition type (it has to be type "DE", which
is in that sense a non-standard partition, but it's really a FAT partition
in disguise).
Bill in Co
see the HPA stuff. I must have missed that page before! Thanks.
Bill in Co
It could have been possible, but since chkdsk says it's ok, I don't think
that's it. Besides which, modern HD have ways to prevent that from
happening from what I've read at one of their sites (meaning, they test it
for such conditions).
UPDATE:
Nope, don't have a HPA, so at least that is my favor!
I still find it a bit incredulous that ATI was seemingly able to clone all
the drive and all its partitions, but that now the Ctrl-F11 option (for Dell
System Restore capability) is not able to run the Dell Restore code linked
to the third partition (it goes there at bootup after Ctrl-F11, but locks
up). I went over to the Dell site and apparently this issue has been seen
before. I think the only solution (if I want to "go there") is to do a
time consuming, sector-by-sector clone, instead of the normal default of
only copying sectors with data in them (as recognized by the cloning
software - and that may be the catch, come to think of it).
> Thanks Paul, and I'm well aware of Goodell's site. :-) That's where I
> got
> the DSRFIX (Dell System Restore) fix utility from.
> The interesting thing is, everything is fine here unless I try using that
> Ctrl-F11 bootup option (to allow access to the Dell system restore option
> if
> one wanted it), and if I use Ctrl-F11, it ends up locking up the system,
> so
> I then need to reboot and run DSRFIX on a floppy so that I can ever boot
> up
> again. The active windows partition here is #2, I think, since at
> bootup,
> Dell's hidden partition initiates everything from startup to go through it
> first before it passes control over to windows.
>
> But under no circumstances am I ever able to successfully run the Ctrl-F11
> restore option anymore (it fails and locks up the system at the BIOS
> level,
> and then I have to "fix it" with DSRFIX (Goodell's Dell System Restore
> utility fixer), which at least allows me to reboot into windows again!).
>
> Clearly SOME code was NOT cloned over from the original drive using
> Acronis
> True Image's cloning option. I'll probably go back and restudy Goodell's
> site again, but I think this is irreparable without editing some disk
> sectors manually, and I don't want to (or know enough to) "go there". At
> least I knew enough to use DSRFIX to be able to reboot again after trying
> Ctrl-F11 which messed up the partition type (it has to be type "DE", which
> is in that sense a non-standard partition, but it's really a FAT32
> partition,
> in disguise).
Paul
> UPDATED (for anyone who's still following...)
>>>> I could have chosen to use the (much more time intensive)
>>>> sector-by-sector
>>>> cloning process, but decided against it, figuring that I didn't need to
>>>> do that. And I don't, so long as I give up the Dell Restore to factory
>>>> setup fallback capability.
Any cloning utility, can only "clone what it sees". If the computer
BIOS does things to hide some of the sectors, even a sector by sector
approach will fail. An HPA is just one example of a trick that
hides the sectors. Dynamically fooling around with the MBR
(at the BIOS level), is another potential trick.
You'd need to find a second computer, sans "tricks", to successfully
copy all the sectors. That's why I'm claiming there is a
"challenge" to this. It isn't a slam dunk.
Paul
Bill in Co
> Bill in Co wrote:
>> UPDATED (for anyone who's still following...)
>
>>>>> I could have chosen to use the (much more time intensive)
>>>>> sector-by-sector
>>>>> cloning process, but decided against it, figuring that I didn't need
>>>>> to
>>>>> do that. And I don't, so long as I give up the Dell Restore to
>>>>> factory
>>>>> setup fallback capability.
>
> Any cloning utility, can only "clone what it sees". If the computer
> BIOS does things to hide some of the sectors, even a sector by sector
> approach will fail. An HPA is just one example of a trick that
> hides the sectors. Dynamically fooling around with the MBR
> (at the BIOS level), is another potential trick.
And I think the Dell does that - changes the partition "type" as its boots
up, depending on what keys are or are not pressed. For example, if you
press Ctrl-F11, it redirects to its Dell Restore option, instead of Windows.
(But if that fails to be successfully found, you're screwed, unless you run
that Goodell DSRFIX to reset the partition type to get back access to the
windows partition (BTDT). (Sloppy coding on Dell's part, for the case of
when it fails (like after a cloning operation), and then you can't even boot
up anymore).
> You'd need to find a second computer, sans "tricks", to successfully
> copy all the sectors. That's why I'm claiming there is a
> "challenge" to this. It isn't a slam dunk.
>
> Paul
Very interesting, Paul.
I guess by "using a second computer", it's the fact that it is THEN not
relying on anything on that connected hard drive, as though it were just a
dead appendage. So presumably there would then be (pretty much) unlimited
access to copying sectors, without any interference from BIOS? Is that
it?
But I'm still a bit mystified as to how the BIOS can (or would even want to
try to) "hide" sectors, from a basic sector-by-sector clone operation.
Perhaps you mean that in the partition table(?), or in the MBR(?), it
somehow can mark some sectors in some way such that a cloning program isn't
even aware that these sectors are in use?
Paul
Perhaps you should read the section on HPA first. This will give
you some background info. I've never worked on one of these,
so I have no idea how tough it is to work with.
http://en.wikipedia.org/wiki/Host_protected_area
When I referred to a second computer, I was referring to the possibility
of finding a computer that wouldn't prevent access to the HPA. So that
you could extend the drive to full size, use something like TestDisk to
scan for partitions, find out where the (hidden) partition starts,
figure out a way to mount it, then back it up. That would be
a challenging project. The machine I'd reach for here, would be
the oldest computer that supports 48 bit LBA (in case the disk was
bigger than 137GB). If the disk was a smaller one, I could use
my ten year old 440BX based computer, as it might not know what
an HPA is, and might not be taking any precautions to prevent
me from getting at it.
Paul
Bill in Co
I looked over that article, but I still don't clearly see how a low level
sector by sector copy can't work - except that the reported disk size might
be off.
It just seems to me that a sector-by-sector transfer is practically at the
hardware level. That is UNLIKE anything that needs cluster or file
location information, or knowing where specific data is located on the disk,
as indexed by the MBR.
IOW, a sector-by-sector disk copy operation is a low level "dumb" sector
copying operation, so how can it miss anything (except (possibly) due to a
reported disk size "discrepancy", where it wouldn't catch some of them, in
some seemingly hidden region of the disk)? Is that just it?
Paul
I think the idea of the HPA is, you can't access the upper part of the
disk, when it's in place. That's why a sector copy isn't going to help.
All those addresses are effectively "past the end of the disk", when
you go to try it.
Maybe it's time for me to try my hand at making an HPA, using one of
my older computers. That sounds like a fun project.
Paul
Bill in Co
OK, I get that, but this is a pretty special case. AFAIK, I don't have a
HPA in my case, since I don't have a Dell Media Direct utility (which is
apparently installed on notebooks), AND the fact that the two special hidden
Dell partitions were, in fact, copied over to the clone (although who knows
if ALL of their contents were).
Hot-Text
will always need new idols if it work out or not!
Thant you
Hot-text
saikat acharyya
any bad sector or u can use in cmd prompt.where u have to type chdsk c:
then press enter.thats all
Submitted via EggHeadCafe
Visual Studio Async CTP Overview
http://www.eggheadcafe.com/tutorials/aspnet/e78665f9-b949-4cc7-a024-654eb3bbad6d/visual-studio-async-ctp-overview.aspx