Jo-Anne wrote:
SC Tom wrote:
The ancient hard drive in my WinXP computer is probably dying. I
have a new drive to install. However, I was hoping to get one
more
complete backup/clone last night with Acronis True Image. The
backup
to an external USB drive didn't finish, however; and the screen
wouldn't unblank. I finally unplugged the drive, turned off the
computer, and turned it back on. It came on OK and is working OK.
I thought I'd do a backup of at least my datafiles onto a
DVD--but
the program I was using found errors (unspecified) and couldn't
finish. I do have a clone from 4 days ago, and I've backed up the
most important datafiles, Outlook Express, and Favorites.
Should I try running CHKDSK/R? Or is there something else I
should
do instead that might allow me to make one more backup/clone?
Thank you!
Jo-Anne
Did you run True Image from within XP, or have you created the
boot
disk and run it from the CD? I've had better results with flaky
disks
running it from the CD, but I run it that way anyhow since it's
faster, and there's no chance of a file being in use.
After following Paul's excellent advice, I'd boot from the ATI CD
and
create an image that way.
--
SC Tom
Thank you, SC Tom! I did run True Image from within XP. Tonight
I'll
try running it from the boot disk. (I did that only once, a long
time
ago, to make sure the boot disk was working and I could do the
backup.)
Jo-Anne
With the disk not being in use except for access by ATI, I think you
may have better results that way.
In response to your reply to Paul, I wouldn't think 39C (102F) is
too
high. My EIDE drive runs a steady 40C (104F) and my SATA 48C (118F)
and
I've never been warned that those temps were too high, and have had
no
problems (KOW!) for the last 3 years I've had them.
It can be a combination of high room humidity and high temperature
that can cause drive failures. If you're in an air conditioned room,
that keeps both the humidity and the air temperature down, which
is "twice as good". Hard drives have a "breather hole", and moist
room air can eventually affect the environment around the platter.
The filter on the breather hole, keeps out particulate, but if
any molecules are small enough, they get through.
I've had one drive failure here, during a period of time where the
room humidity wasn't controlled (failure after about one month under
those conditions). The manufacturer documentation sometimes contains
a temperature/humidity graph, showing the acceptable range. If the
room is humid enough, even 40C can be too hot.
Paul
Thank you, Paul! The room is air conditioned, and the humidity isn't
terrible--although the Midwest this summer has been VERY humid indeed.
Did you see my two responses to your first response? In case not,
here's
the gist of them:
1. I ran HD Tunes, and Error Scan showed only one "damaged" square (22
MB) for the whole drive. Health showed a yellow highlight across
Reallocated Sector Count, but the Status for everything was OK. I
don't
know if Benchmark and Info are useful in this context. I can report
back
if anything is important.
2. Today I tried backing up my datafiles to DVDs. Everything was OK
except for some folders containing photos. The Easy CD Creator error
messages after the latest failed backup are
E80041898: Write error - [03/0C/00]
E80041925: TrackWriter error - Command retry failed - [T7118]
I'm not sure what this means, and the program wasn't forthcoming with
info.
3. Addendum: Later in the day I was able to back up all the photo
folders--one folder per DVD.
Jo-Anne
So then it sounds like you rescued all of it ? Did the software you
used,
log or record that all files have been captured ? That is what I'd want
to know at this point.
You should be able to restore that to a new drive.
In any case, I wouldn't erase the old drive just yet. The thing about
making optical media backups, is you may be in for a rude shock when
restoring from them.
The two error messages are just one error, a write error. The 03/0C/00
part of that error message, seem to be relatively constant from one
report
of that to another (looking at various user reports of that error
message).
You can see some suggestions here about what to do.
http://www.cdrlabs.com/forums/anyone-have-samsung-drive-need-some-help-t10414.html
With the optical drives I've got, I usually take care of the firmware
updating
first, before starting to use the drive for serious work. I like to
burn
test
discs, then scan with a Nero diagnostic, to determine whether the
burner
is
doing a good job or not.
(An optical disc, scanned after being burned. Error rate is low and
looks
good.
If the graph gets into 1000-10000 range, the disc is probably toast.)
http://gfx.cdfreaks.com/reviews/lg_gh22lp20/image138.png
There is one utility, besides this one, which can check out burned
discs,
but it is supposed to work with Liteon drives (program named KProbe).
The utility in this picture, is available for download, so you don't
have
to buy it. The author of this tool, also makes it available for
download
separately.
http://gfx.cdfreaks.com/reviews/lg_gh22lp20/image134.png
The thing about optical media, is it has three dimensions of error
detection
and correction. It allows, for example, a scratch to be present on the
optical
disc, and still recover the data. But if the media burning process is
flawed,
the errors end up in the 1000-10000 range, while a good burn might be
around 10
or so (there are always errors, so some errors are to be expected). If,
when
you make a backup, the graph of errors over the surface, have a rising
characteristic, and get worse near the end of the scan, that can have
consequences when you go to do a restore (it won't finish). That's why,
if
you think you're making archival copies or "only" copies of computer
data,
at the very least you want to scan the disc to see if it is already
compromised.
If the burn is showing a high error rate, in the 1000-10000 range, then
it is quite possible that CD or DVD is going to fail in a couple months
of sitting on a shelf.
If I have a spindle of discs I just bought, I might scan one or two of
them,
to see if they're OK or not. For media where I've had nothing but
problems,
I have to scan every one of them. So the scanning process is
"adaptive",
in the sense that you scan every piece of media, if you suspect the
media is bad stuff.
So an error message, when attempting to burn files, tells you that you
lost
something. There can be trivial errors when burning though. I've had
problems
before, where the "verify" phase of a burn (verify turned on in the
burning
program) fails reliably, every time, and yet there is nothing really
wrong
with the information content. If I make an ISO9660 file from the media
contents
after the burn, it appears to be perfect, with respect to having
reproduced the
original content. So I have had occasions where the verify just doesn't
work
properly, but there is nothing really wrong with the discs. But if you
have
a failure in the middle of a burn, or a later scan of the media shows a
high background error rate, these are not good signs, and you could be
in trouble immediately with that disc, or later when you need it most.
*******
The fact you can see one colored block in the HDTune result, tells you
there were some sectors that could not be spared out properly. Or some
sectors that have gone bad since they were written.
So your primary task right now, is to make sure all the information on
that hard drive is safe. You could eventually attempt to erase the
old drive, and then an HDTune scan may report all is well. (The SMART
statistics may still be bad of course.) If after erasing the drive, a
scan still shows errors, then it would mean there is a bad spot on the
disk, for which there aren't enough spare sectors to fix it.
On modern drives, doing "format" to the disk, doesn't really do a
complete
restoration of the surface. Modern disks use embedded servo, meaning
there
are bits of information recorded between sectors, that cannot be
re-written or refreshed by the user at home. The servo pattern is
written
by a "server writer" arm, which is poked through a hole in the side of
the
hard drive at the factory. If you look at some of your 3-5 year old
drives,
you may see a silver sticker on the side of the drive, in the platter
area.
They shove a "head on an arm" through that hole, and the servo writer
establishes the embedded servo pattern. When you do a "format" on such
a
drive, it may rewrite some portion of each sector, but it can't refresh
the servo recording. The format gives an opportunity for any "pending"
sectors to be spared, if they've already been detected as being bad.
The format may give a false sense of security, if the HDTune scan looks
clean after you're finished. A format (or just writing to each sector),
has about the same value in this case. It's a "feel good" operation.
The highest density drives now, no longer use that crude "head on an
arm",
so the "port" on the side of 1TB to 2TB drives will have gone missing.
The track density of modern drives is so high, it isn't possible
mechanically,
to get an accurate servo pattern using the old method. The drive
itself,
now helps with the servo writing process (I haven't seen any details as
to how the arm is held in relatively the correct position, during that
process). It doesn't imply though, that a 1TB to 2TB drive is going to
rewrite embedded servo information, just because they can do it that
way. There may still be factory equipment, controlling the process,
so the solution on modern drives may not be self-sufficient enough to
just rewrite the whole surface. (It would be nice if it did.) It just
means the servo writer port, no longer exists, so there is one less
silver sticker on the side of the hard drive. (The fact there is a
silver sticker present, tells you the HDA holds neither a vacuum
nor is pressurized. The sticker seems to use good adhesive, and
I've not heard of the sticker falling off.)
On much older disks, a format wrote more of the pattern on the disk.
We used to have soft sectors on hard drives at one time, and doing
a "format" on those, really re-wrote the entire surface. The reason
that was possible, is all the "servo" pattern was written to one side
of a platter reserved for that function. (You might have four platters,
eight recording surfaces, seven surfaces for data, and one surface for
servo pattern only.) So formatting in that case at least, was
attempting
to write *everything* on seven of the surfaces. But if you have a
platter
with a bad spot now, formatting or erasing the sectors really isn't
doing anything other than giving re-allocation a chance to attempt
to spare out bad sectors.
It's one reason, if I see much in the way of complaints, either in
SMART, or in an error scan, I replace the drive immediately. A
"bad spot" in a hard drive can be physical in nature, such as a
ding in the surface, debris floating around inside the HDA or the like.
The drive has some air filter packets, which remove "stuff" ground
off the platters, but if the surfaces of the platters are failing,
you'd find the filters are filthy and there is junk everywhere.
The atmosphere inside the drive is supposed to remain super-clean,
but when the plating on the platter is failing, is can be
pretty dirty looking inside there. And easy to see why there
are so many bad sectors.
It's too bad more discs don't have transparent covers, so
you can look inside and see how they're doing
I had one
drive at work, that I was developing a controller for, that
I was actually able to see a malfunction happening, because
it had a transparent plastic cover over the platters.
Those were the good old days
In modern times, there
was one version of Raptor drive, that comes with a
transparent window on the drive, and that is about as
close as we get now, to seeing how it's going.
Paul
