(Long ...)
[]
"J. P. Gilliver (John)" wrote in message
Is there anything that will read a file with a dud block (on the hard
disc) in it, replacing the dud block with zeroes, ones, or something?
I don't mean something like SpinRite, that tries to reread the dud block
up to 2000 times in the hope one of them will succeed; I just mean
something that'll copy _past_ the unreadable block.
A few things - like IrfanView - will read _up to_ the dud block, and
give you the first part, filling the rest with grey; most things, such
as Windows Explorer, just read up to the dud block, then pause for ages,
then give you an error message without giving _any_ of the corrupted file.
I have a few videos I'd like to recover from the dud disc - with maybe a
few dropped frames.
Ideally something free and GUI, rather than paid or command line: I just
want, ideally, to be able to drag the corrupted file to a new location,
and get a file (with some dud blocks) that at least can be read without
lockup-and-loss.
Just thought I'd give you all a summary.
I did have: Samsung NC-20 netbook (a 12" one, so the keyboard has almost
full-size keys), with a (Samsung) 160G drive in it (as originally
supplied). XP, my main machine.
Acer Aspire 9301, 7HP. (17".) Bought _mainly_ to support (including
several blind) friends who have 7, but also to get more familiar with 7,
and also to have the power for when I need it (it's a much more modern
PC). [320G drive.] (I rarely do need it: I'm not a gamer, or even much
of a video watcher. Mainly email, usenet, and genealogy.)
I'd always backed up important data to a second partition on the HD,
since in the past my experience in the past has always been that HDs
give some indication of impending failure, at least long enough to copy
data off. However, I always "meant to" get a physical backup drive, not
only to back up to a truly separate drive, but also to use one of the
things we discuss here occasionally - Acronis, Macrium, etc. - to back
up the _system_ area so I can restore my XP system if I need to. (The
intricacies of how XP OEM [the netbook came with XP SP3 preinstalled]
validates/authorises itself having been something I've never really been
that interested in learning.)
Then - the _day_ before I was due to go to a computer fair at which I
was going to buy my backup drive (actually probably only a few hours
before; I play late), the drive on the netbook stopped - suddenly, with
no warning.
I went to the fair anyway, and bought a couple of drives (a 1TB 3.5"
purely for backup, and a 250G 2.5" for the netbook). Both with 3 year
guarantees - one can't be sure of anything these days, but I thought
that was my best bet. (If anyone's interested, the 1T is a Seagate and
the 250G is a Hitachi.) (I also bought an external dock - red and black
thing, does SATA and PATA [and USB, and card slots ...]: the quite
meaty-looking "3A" 12 power supply it came with died after a few days,
though fortunately I have another one that's serving. A friend tells me
he's seen on t'internet that the PSUs with these docks are commonly
failing, in one case fierily so. I'll take it back next time I go to the
same fair.)
During the next few days, I tried the various methods on the internet
for recovering a dead drive, including the freezer. I could tell (hear)
it wasn't rotating. (Actually consensus is that the freezing is _not_ a
good idea because of condensation on the platters causing head crashes
and damage when it does spin up, and I think there may be _some_
validity to that - it wasn't important in my case as it _didn't_ spin
up. My trust in the people who warn about condensation wasn't boosted by
one of them - on YouTube - _showing_ the condensation; obviously, to do
so, he'd opened it up and exposed it to the air. Others claim that this
isn't too relevant as they're not actually hermetically sealed anyway;
OK, and yes mine has a breather hole, but I'm still not sure that's the
same as actually opening it up when its cold.) Another suggestion I
found on the 'net was that the overvoltage protection diodes sometimes
fail _short_ circuit and thus prevent power reaching the drive (or part
thereof); this seems unlikely to me (such diodes usually IME - and I
work in avionics, and one of the units I service is always coming in
with them blown - fail _open_ circuit), but lots of people said it had
worked for them, so I disconnected it (at one end): no change. (I
haven't got round to reconnecting it.)
Even considered using a data recovery service. But I had concerns about
whether they'd keep copies of what was on the drive; however reputable
the company, it only takes one employee ... In practice I'm glad I
didn't, as I don't think they'd have done _much_ better than I did in
the end.
So, eventually, I bit the bullet, and - in our company clean-air cabinet
(incidentally: such cabinets, and clean rooms, operate under _positive_
pressure, not what you'd intuitively think), after running its fans for
a while - I opened up the drive. (A Torx #6 'driver fitted the seven
screws; note there's one under one of the labels, but unlike some of
those on YouTube, I didn't poke through the label, but was able to
carefully peel it back, and replace it afterwards.) I found the head on
the drive, not parked to the side as it should have been. I very
carefully turned the drive - using the driver; the same one fitted the
hub. I felt slight resistance; the head(s?) had indeed obviously stuck
to the drive. I carefully (without touching the platter surfaces) moved
the head assembly back to its park position, reassembled, took it over
to my dock, and tried power - whee, it span up. Rushed home to extract
as much as possible before the doom set in predicted by all those on the
net who said never open a drive. (To be honest, I suspect that - for the
short time involved, at least - opening in a normal home, as long as
there are no smokers or similar - you'd get away with it anyway;
especially if you made your own clean cabinet with a fan and some
filters [positive pressure, remember]. But don't quote me on that!)
Anyway, I got home, and set to _moving_ - not copying - from the two
partitions, to a couple of folders (called something like C-saved and
D-saved) on the other laptop. (Couldn't use the dock to copy straight to
the new backup drive: it only has one SATA slot!) Now, before you all
suck in breath at my doing a move rather than a copy: I'd tentatively
looked first, and saw that the drive appeared to be mostly OK. Doing a
move meant that I could _easily_ see what hadn't copied.
During subsequent days I tried to recover the few files that didn't come
off the first time. I think I might have recovered one or two.
When I'd _more or less_ given up (I still have the dud drive), I
Macriumed an image of the recovery partition (100M, IIRR) and the C:
(about 30G). This was mainly to give myself the best chance of getting a
valid XP system back.
I then put the new (250G) drive into the dock. First, I restored from
the image, so that the recovery partition and boot parts of C: were
"restored". I then resized C: to 40G, made a D: partition for the rest
(I still intend to keep most of my data away from C:, to [a] keep the
image size down,
keep _data_ backing up something I can do with
plain copying), and then moved files from the two directories on the 7
laptop back to the two partitions.
(I spent a few more days trying again to get the few dud files back.
Without success.)
A couple of days ago, I finally bit the bullet again, put the new drive
into the netbook, and powered it up. Something Samsung came up: it
offered me at least two options, one which was a minimum restore (or
repair or something), one a full (or something like that) which it said
would erase everything. Obviously, I chose the first one. After a little
while, my system is back!
After a very few more repairs, obviously first thing done has been make
a new Macrium image - now of the hidden partition and the (resized and)
reloaded C: - and a copy of D:, this time onto the new backup drive.
(I'd made - and used - the Macrium boot CDs [two, one for 32 bit and one
for 64; not sure whether I needed both, but I got the impression I did];
incidentally, they fit onto mini-CDs.)
So all is well, and I'm back typing on the netbook. (The 17" one is a
bit heavy on the knees!) [It's _so_ nice to be back using the software
I'm used to (which won't run under 64-bit), not to mention have all my
back emails/news for many years!]
The one remaining thing I'd _like_ to do is get back the remaining dud
files: in particular a few videos I'd taken, which would probably just
show a glitch for the corrupted sectors. I've kept several of your
suggestions in this thread, and may pursue some of them at some time in
the future, when I feel up to doing battle with Linux and/or command
line. I must say I've completely given up any hope of reading the bad
sectors, so please _don't_ suggest anything for that; however, if anyone
does know of a free, and Windows (XP or 7), utility that will read a
FILE that has a few (I suspect only one for most of my files),
zero-filling (or whatever) where the dud bits are, I'd be very grateful.
(In the distant past, under BBC BASIC in the days of floppy discs, I
even wrote something: it read a byte at a time, wrote it to another
file, and closed the other file in the event of an error. That at least
gave me files up to the dud, though not the bit after.)
When I look at the dud disc in Windows explorer (under 7 at least), and
look at a folder with images or videos in, it shows me thumbnails for
most of them, which is at least partly why I suspect they're mostly
single sector faults. I also scanned the whole drive with HDDScan, which
gives a nice graphical picture as it goes, and didn't see _any_ adjacent
bad blocks, only single ones (though I could have missed a _few_ - it
did take some hours!). For interest, HDDScan gave the following totals
(access times):
< 5 ms 112703 sectors
<10 ms 1058776 (probably would have been <5 if not for USB, multi, &c.)
<20 ms 7049
<50 ms 18991
<150 ms 11378
<500 ms 4781
Bads 5297
I presume the Bads (less than half a percent, though obviously enough to
consider the drive scrap) are from where the head(s) stuck to it, plus
_possibly_ - though I was careful [and this is where I think a data
recovery company could _possibly_ have bettered] - from my moving them
back to the park area. (They were about half way across the disc; do
modern discs start from the middle or the edge?)
Oh, and a final thought/question: what caused them to stick? Also, I've
just checked my Power Options: I'd thought maybe I should change from
turn off hard discs to Never, but I see I already have it set to Never
anyway when on external power, which it was when it died (I thought
maybe I'd got them set to power down). So it seems they must have stuck
while spinning (which, on reflection, means I was lucky there wasn't
_more_ damage). I'm pretty sure they were stuck: initially after failure
when power was applied there was a little tinkling, which a friend
thinks was the heads trying to move, and when I eventually opened it and
turned the spindle, I _think_ I felt a definite "unsticking".
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf
I would have suffered a hell of a
lot more if I had been understood. -Clarence Darrow, lawyer and author
(1857-1938)
%%
I hope you dream a pig.
%