Every time reading certain sectors the drive just shuts down:
Then cant remount it. even plugging the power of the drive and reattaching wont help. I need to reboot the system every time..
i found more than 100 sectors at completely different positions that are making it shut down, it always are the same ones.
the drive approaches some of these sectors with full speed of 100mb/s for others it slows down to 0.1mb/s a minute before it stops. It runs all quit before it just shuts down, trying to make a complete image its all quiet. no repeated clicking noise over and over like very worn off drivers.

- is this rather a scratch or writer or the whole arm that got imprecise?

Please evaluate possible solutions:
1.Change Firmware - which CC3*?
2.Use a Hardware or Software with "HPA clipping" plus able to power off&restart the drive? which device/software?
3.Force the drive to read all changing settings or modules of the firmware? which?
4.Replace PCB board?
5.consult quality head fix/replacement service ?

Drive:
Barracuda 7200.12 1TB
ST31000528AS
S/N:6VP0HHTQ P/N:98L154-568
Firmware:CC35
Date Code:09465 Site Code: SU
Smart status "replace within 24 hours"

More:
Then drive runs fine until it reaches a certain sector.
An image with DD rescue of 70% took me about 50 hours and has more than 100 gaps as said above
The positions it stops are always the same and the drive seems relatively stable in this state,
BUT i think i also noticed 3 or 4 sectors making it shut down, that came up additionally just during these 50 hours
so here i paused babysitting the drive like that manually since i was afraid it was an endless mission...
The drive was in an external case and made a relatively loud noise at start-up mostly since it was bought new - eventually it ran 6000 hours fine before it stopped.

want the data back (drive does not matter)
And as said i was able to get 70% of the data with dd rescue and rebooting more than 100 times and starting some MB or GB after the sectors it stopped.

Just the fact is annoying, that it does not always keep trying
but just shuts down at damaged sectors and even requires to reboot
it does not make a clicking noise of retrying ever like other old drives
(or is that only when you boot a system but not when you read sequentially to make an image?)

i assume the firmware is "too cautious" here and the platters and the reading works fine but when it notices a certain amount of data corruption it just escapes by a shutdown
Thus I wonder if i can force the firmware by editing settings/modules to just do a straight run of dd rescue one time, no matter what?

-In case there is a serious scratch (dont think so) the drive would react in the same way?

-why is hardware cloning superior when it also only connects to sata and power?

-Here are notes i made in megabyte positions on the disk
of ranges of data i got and positions where it shutdown (the notes are not most accurate because i was hoping it would be fewer of them required)
still maybe they allow some kind of interpretation.
as said before these positions are fixed and will always be the same but during trying for about 50 hours i also notice some new ones i think when i tried positions twice.

Start-Shutdown
0-450
600-2699
-2760
2800-2890
-3500
-3769
-5000
-7705
-11000
-15000
-16000
17000-17329
-20079
-22000
-23000.199
-23100
-25000
26000-26051
27000-27043
-30000
-35000
-38000
-39000
40000-73346
73400-74818
74900-75036
-76838
78900-78997
-78999
-87752
-87800
-87805
8?500-157000
157500-163400
-165000
166000-321013
-330099
-340091
345000-398000
405000-
415000-
-444544
450000-458168
458180-49????
-498189
-52396*
530000-533531
56*000-587131
-598191
-605702
630000-633117
639900-662210
662300-
662700-
663900-664339
-669673
669900-670442
680000-680428
680450-685566
-685785
688000-688758
-689034
-690278
738567-738663
739000-739***
-750000
751000-751172
760000-760832
(will add the rest if anyone is interested )

Agree with Spildit, get it to someone with the proper hardware to clone this drive.

AT THIS STAGE it should not be too expensive, but if you "DIY it to death" then you could be facing a VERY expensive recovery bill

_________________
PC Image Data Recovery
http://www.pcimage.co.uk

New!! HDD-PCB.COM for all your PCB and donor HDD requirements!

I've got a 7200.4 doing the same thing

Reading certain sectors causes the drive to go totally offline / blocks / goes not ready

power cycling the drive usually gets it back ready again
(but sometimes my disk tools, like Victoria still cant see it until I reboot the tools system)

I eventually (repeated software imaging at pre-calculated offsets) managed to get most of the areas I needed off

Was thinking of seeing if the Seagate terminal could read the problem sectors and what it said about them

And I also have the question
"-why is hardware cloning superior when it also only connects to sata and power?"
All it can do is extra and repeated power resets
Its still using standard read commands (or maybe read-longs?)

hardware imagers are designed with failing drives in mind: on encountering a weak sector the jump forward (or can work in reverse) away from the damage trying at first pass to read all good sectors. Many fore passes will occure- jumps will be smaller and small until last pass bad sectors only are read. Automatic power cycling is built in. The operator can reverse processes and interact in real time with what areas are being copied. Using software imagers the drive will work too hard on bad areas before the readable data is copied.

Pretty much so, yes.

But also with hardware and professional software solutions we can do many other things, including different reading methods (various UDMA/dma/pio modes, initialise read and so on), we can also switch off background tasks like auto reallocate etc...etc..etc..

_________________
PC Image Data Recovery
http://www.pcimage.co.uk

New!! HDD-PCB.COM for all your PCB and donor HDD requirements!

Many thanks for your answers!:)
still wondering what is the actual failure of my drive, corrupted data only or hardware failure?
you don't think it can be fixed by editing the firmware?

i would at least like to try one of these "shutdown sections" again now for several times,
while mounted read only and dd rescue (what are names of more sophisticated software?)
and while changing every setting that makes any sense to see if it makes it go any better.
- like: and connecting to the firmware with a usb or com ttl adapter

Are you willing to point out roughly what settings i will aim to change?(/which of those are most promising?
where to find them ?/ which i wont be able to change?)


-Which hardware cloner or professional software to choose for my drive and why?
-What's next if the scenario happens that the hardwarecloner's log will show 1000 sectors where it switches off
and have 90% of the data after running 10 days but i wanna go for 99%?


-which two price ranges/repairs do you mean?

In short again,
-does anyone have a clue whats my drives hardware problem (if any)?

-Whats the minimum requirements just to try to change any of these modes or more manually on a slow/shutdown Section of the drive (Can't be that all this can only be done by hardware cloners?)

-How much more data will any hardware disk cloner like deepspar get, which one is the best to buy for this case?
I can choose a minimum speed and read in reverse in dd rescue too, still it will hit many Sections to shutdown..
and can i tell the cloning machine to write on the same receiver drive i started with and skip the 70% i have already?

here is the rest of my notes (just in case)
770000-770004
800000-
810000-8145508
-815090
825000-826334
826500-826702
827750-843727
844000-844134
876-
-880260
900000-911000
-911995
-920402
922000-924291
924400-924966
925000-927923
928000-
930000-943149
951000-
953000-
954000-954418.974
955000-955020.905
965000-965410.660
970000-970130.172
975000-975100.075

Thank you=)

Where to find a full list of VSC / Vendor specific ATA commands (for this 7200.12 / Seagate "F3") ?

yes the hardware imagers add to what is already imaged without repeating already imaged sectors. Advantage to hardware is that power cycleing can be an automated process: "on 10 bad sectros- then power cycle and resume"

a simple shell script can do reboots (for hdd power cycles&detection) and resume of ddrescue, a minified linux on a ssd can boot very quickly
also will try differnt udma/pio modes by hdparm.
not sure yet how to switch off auto reallocate and more things tho, which "pcimage" mentioned

Asked for the Vendor specific ATA commands because i wanted commands to pause single heads for example?
In case only one head is causeing the shutdowns / has bad areas, to get 100% data of the other heads first.

is there is more then already said here that a hardware imager does?
s there more that someone good with ATA and Segate Terminal (or pc3000) would try besides analysis?
Thanks!

Or with a bit of electronics (I used MOSFETs with over-current detection, controlled via USB commands), then power to the drive can be switched without rebooting the PC. There are many possible options...

yes @vulcan just this drive cant be remounted without rebooting the computer as well

only still remaining questions yet are:
-if there is any chance to find/obtain the vendor specific ATA commands to disable certain heads while cloning or more
-and if the hardware cloners do any more standard or special things
(@pcimage how to "turn off auto reallocate" in linux?)

thanks:)

@ buhlau experiment live on patient are not reccomanded in your case, now recovery with proper tools and knowledgment can be a low price, if heads start to fail and SA become more demaged ( yes, if you power OFF/ON drive the weak/demaged haeds need to read again in the same service area ) than you will need to duplicate the recovery fee or get less data.IMO.