Hello there! This is my first post here and I'm a total beginner, please be kind

I have this old WD External HDD, which, until today wasn't detected at all by my computer:
https://www.amazon.com/Passport-Essenti ... 002OB49SC/

This is the sad type of drives that not only is missing a SATA port, but also has some hardware encryption built in. I opened the enclosure to have a look at the PCB (here are some HD images) and long story short, the connection pins on the right of image 1 were quite tarnished and I cleaned them up with some light abrasion and isopropyl alcohol (before that the drive's LED would just flash indefinitely).

The new situation is that the drive is now seen by my OS, but unavailable still, and throwing a bunch of IO errors (refer to ). This drive normally emulates a CDROM that comes preloaded with some software but even that doesn't show up. I had a stab at smartmontools (with the result in ). This all looks super alarming, but the drive itself sounds healthy (I could try to take a recording of it, but there is no clicking and it hums gently).

Now I would need advice as to what I should do next. Some guides recommend swapping the PCB which is a rabbit hole of its own: the firmware chip of the current PCB would need to be transferred to a "donor" PCB (to preserve encryption is my guess), which involves some precision soldering which I'm under-equipped to perform (but am willing to adventure into would my chances be high).

I think that would only make sense once:
1- I make sure that the drive itself is healthy: do you see any indication of that (or its opposite) from the logs I attached? …and
2- I make sure that the firmware itself isn't at fault, is there a way to check that? Alternatively, there seem to be ways of unencrypting the device in software (which would spare me the soldering step), but is that possible from Linux?

Thanks in advance and sorry if all that sounds naive, as I said, this is pretty new to me

if drive itself sounds healthy, so most likely you don't need to change pcb.
Do you need data or you want to play data recovery game?

hey drHDD, thanks for picking up this thread

This drive has been an itch I've been wanting to scratch for years. I value some of the data that's on it (and of which I have no backup), but it's been so long that I wouldn't hate myself too much for messing it up as long as I did my best in the process.

The drive does sound healthy (I could perhaps attach a recording of it spinning up next). When you say that the PCB probably doesn't need to be changed, what do you believe the likely cause to be, then?

Just a quick peek and looks like it's suffering from bad blocks.
There's no need to change the PCB and ROM, but if you necessarily want to change the PCB to use a SATA interface instead of USB, this should be a suitable PCB: https://tinyurl.com/6tssf6fk
Sometimes using a USB interface might be painfully slow compared to SATA.

Maybe the best free IMG tool that can handle bad blocks is HDDSuperclone for Linux which has been mentioned quite many times here.

I would say your chance to get the data should be good and the DR lab's chance to get the data should be high, only you decide the path you want to take.

Good luck!

Sorry for the slow response and thanks very much for the info & encouragements!

I had a stab at HDDSuperClone, it does detect the disk, but here is what running an Extended Analyze returns: ; doesn't look too encouraging

Could it be that HDDSuperClone is fooled by the encryption chip ?

100% bad sectors sounds a bit dramatic for a drive which doesn't sound unhealthy to my hears, here's a recording of it spinning up: https://we.tl/t-5Afi8NykLq (sorry I couldn't attach the file here, it's not allowed by the forum).

What do you think should be my next step?

from simple to complicated:
( IO errors are from bad/loose/dirty usb cable, loose/dirty machine side usb connector.
loose/dirty disk side usb connector, corroded controller board connector to disk )
-try different ubs cable, cheap thin chinese cables are crap
-clean all usb computer/laptop contacts as well, use air blow cans, fine screwdriver or micro brush with alcohol to rub them a bit, make sure usb connectors are tight
- remove controller board from disk, clean controller_board-to-disk pcb contacts ( visible on the right side of your pictures ) with rubber eraser until they shine well, use alcohol prep wipes to clean all contacts on controller board even gold plated ones + blow air on the board.
-connect disk to win 10 machine ( avoid linux for now ), see if disk is:
a. visible in disk management
( ***IF PROMPTED TO INITIALIZE DISK - DO NOT !!!*** )
b. shows as healthy partition vs raw no partition
c. shows online vs offline, try toggling ONLINE if offline
d. shows letter assigned to it vs no letter
if all 4 are there then
win file explorer -> right click on disk -> properties -> tools -> error checking -> check for errors ( fast way ) -> win will check/fix for errors -> win will then recognize it -> boot to linux and it will work

else see which step you get stuck on and post back

Hi, and thanks for jumping in!

- I have a cable which I trust
- I blew some air and scrubbed some cotton in the USB connectors, but since this is micro usb, I was rather worried about making things worse
- I cleaned the pcb contacts to the disc, they were oxidized and now much shinier

About outcomes, I found a W11 box on which the drive eventually shows up after long minutes during which the disk management UI freezes. The disk appears as uninitialized but online, it has an icon showing a red arrow pointing downwards. The good news is that it shows some info like the disk name "WD My Passport 070A USB Device" and its total size. Now regarding its partitions, I suspect the disk has a single ext4 partition and Windows just shows a single slab of "unallocated" space. I'm trying "Diskinternals Linux Reader" to see if it manages to list it/its content, but it spins endlessly in "Detecting disks in the system". On the linux machine, I'm still getting the same IO Error and hddsuperclone analyze still shows as many errors.

I'm just working on the same type of drive WD Passport external, older 2016 drive 1tb
( I'm not a pro at data recovery )
What happened - in linux, connected drive to usb port, drive got mounted ok, started copying files from it ( reading ) , all fine, all of the sudden copy operation stopped and could not progress further in linux, switched to windows.
Connected drive to win 10 machine - drive slow to be detected, slows down windows to a crawl, light blinking.

First I thought it was usb connector problem,
cleaned controller board contacts, checked usb connector to board for continuity all good, checked on brand new cable all good.

Managed to mount, updated firmware - that made drive detection faster but one time only.

Sometimes when connecting under win makes it show files in explore, but then slows down to a crawl,
light blinking mostly, drive is spinning, but most of the time mounting in win is crawling-slow.

Removing 2 drivers ( 2nd one is WD SES driver ) in device manager makes the drive to mount a bit faster, on fresh driver reinstall, all drivers are up to date.

Most of the time shows up as raw/unallocated in win disk management.

One time was able to run crystadiskinfo on the drive - CAUTION status - the only yellow line - 05 relocated sectors raw value is 126, the rest of the lines are
all blue.

If mount in win successful win prompts for scan to fix errors, but error checking then fails with message - read failure with status ... at offset ...

Seatools Fixall ( short and long ) both fail on the drive.

Internet research: shows passport drives slowing down to bogging windows down to be a common problem - the cause being
wd drive is constantly attempting to relocate bad sectors OR driver corruption OR firmware corruption.

this link is the most informative but he is not giving the list of commands to execute to stop WD from constantly trying to relocate bad sectors
that's the one you are using hddsupertool

https://nv1t.github.io/blog/i-hate-you-wd

so if you know the commands in linux to clear the drive per that webpage, please post here.


++ pros as I understand run some king of firmware tweak on Passports that makes it resurrect from super slow to normal, then pull data off asap.

++ seems like we got the same problem, slow detection, slow mounting, slow dir and file system access, bogging down OS - ie typical common "SLOW WD Passport drive"

read my previous post and per this link

https://nv1t.github.io/blog/i-hate-you-wd

Quoting the guy:
"This is due to a problem, when the drive gets “stuck” trying to re-locate sectors, or more specific, adding and managing the list of sectors that are “candidates” to be relocated. The WD drives basicaly save specific sectors proactively to be re-located. This makes the drive rather slow.
A fix is to empty this relo-list and disable the re-location feature."
... and then pull data out

so at this point to solve the problem
- figure out how to use this hddsupertool to temporarily disable bad sector relocation, the manual is short
- make image and recover from it

my problems is worse since the disk is near full.
here is the output of the error checking in win 10:

Chkdsk was executed in read/write mode.

Checking file system on F:
Volume label is WD_PASSPORT.

Stage 1: Examining basic file system structure ...
The allocated length 0x895bd000 is not in multiple of 0x10000 for attribute
of type 0x80 and instance tag 0x2.
The allocated length 0x2a000 is not in multiple of 0x10000 for attribute
of type 0x80 and instance tag 0x3.
The allocated length 0x23891000 is not in multiple of 0x10000 for attribute
of type 0x80 and instance tag 0x2.
71424 file records processed. File verification completed.
43 large file records processed. 0 bad file records processed.
Stage 2: Examining file name linkage ...
32 reparse records processed. 72862 index entries processed. Index verification completed.

32 reparse records processed.
Stage 3: Examining security descriptors ...
Cleaning up 12 unused index entries from index $SII of file 0x9.
Cleaning up 12 unused index entries from index $SDH of file 0x9.
Cleaning up 12 unused security descriptors.
Security descriptor verification completed.
720 data files processed. CHKDSK is verifying Usn Journal...
The remaining USN block at offset 0x33fc7e8 in file 0x35d is
less than a page.
Usn Journal verification completed.
Read failure with status 0xc00000a3 at offset 0x1193000 for 0x10000 bytes.
Correcting errors in the Volume Bitmap.
Read failure with status 0xc00000b5 at offset 0x652000 for 0x1000 bytes.
Read failure with status 0xc00000b5 at offset 0x653000 for 0x10000 bytes.
A disk read error occurredc00000b5
Insufficient disk space to fix volume bitmap. CHKDSK aborted. ( this one is probably because disk is near full,
Write failure with status 0xc00000a3 at offset 0x652000 for 0x1000 bytes.
An unspecified error occurred (6e74667363686b2e 1729).


and here is current cmd.exe chkdsk /r /f errors

chkdsk /f /r /x F:
The type of the file system is NTFS.
Cannot lock current drive.
Volume dismounted. All opened handles to this volume are now invalid.
Volume label is WD_PASSPORT.

Stage 1: Examining basic file system structure ...
71424 file records processed.
File verification completed.
43 large file records processed.
0 bad file records processed.

Stage 2: Examining file name linkage ...
32 reparse records processed.
72862 index entries processed.
Index verification completed.
0 unindexed files scanned.
0 unindexed files recovered to lost and found.
32 reparse records processed.

Stage 3: Examining security descriptors ...
Security descriptor verification completed.
720 data files processed.
CHKDSK is verifying Usn Journal...
440 USN bytes processed.
Usn Journal verification completed.

Stage 4: Looking for bad clusters in user file data ...
A disk read error occurredc00000a3
The disk does not have enough space to replace bad clusters
detected in file 42 of name .
A disk read error occurredc00000b5
... it keeps on going trying to repair bad clusters

this likely because disk is near full + files are mostly videos and they are huge about 1gb average

So, to narrow down the question for HDD gurus here,
disk is near full, 1tb size
files are huge 1gb on average
chkdsk reports not enough space on disk to replace bad clusters
looks like disk is stuck in trying to relocate bad cluster files but not enough space left on disk
the question is how to temporarily disable bad clusters relocation on this disk, to pull whatever files out corrupt or not ?

Thanks !

...

Hi and thanks for the pointers,
I don't think I'm as lucky as you (or the "I hate you, WD" blogger) are, I don't think my disk is just "slow" (I have clear IO errors from it), though I did give a shot at the hddsupertools. If you want to know the commands to run, go to VSC (Option 7 from the main menu), you'll find 2 fixes there: Option 4 and 5 ("patch mod 02" and "patch mod 32").

I ran option 4 successfully with no improvement, but option 5 just crashes:



There is an option to read and check rom, which it reports to be good… up to a certain point:



(I don't know if this is good or bad news)

This leaves me back to where I was: what should I try next?

- Is it truly a "bad block" issue like SirForensic suspects? All that HDDSuperClone sees are bad blocks apparently
- Does a PCB swap make any sense at all in this situation?

Here's a dump of the modules, maybe someone with this kind of knowledge could determine if that's the issue?

yeah, I found the link for hddsupertool to clear relo-list that brings drive back to normal speed and plan to give it a try and then use ddrescue or hddsuperclone to pull files out.

https://www.youtube.com/watch?v=9EN1HkJZ81w

Regarding your drive, a clue there is that as I understand ( and I could be wrong, this needs verification ) the controller has onboard flash module where it stores a small partition that has all the software the drive comes with WD Smartware , typically mounts that partition as CDROM. So if that is not mounting like it used to then it might be a problem with controller board ??? Just a clue, this needs verification. Controller boards are around $10 from china on Ebay, you just plug in revision numbers into ebay search and it will spit out the sources, you have to make sure board revision is exactly same and as I understand 2 rom chips need to be moved to replacement, probably at any smartphone repair shop that's a 10 min job.

In general, any cold solder case on the board - can create intermittent component connection to pcb, drive heats up and you got a problem, that's why
they say put the drive in the freezer and then pull data out asap, anecdotal way.

Another option I heard is that to use a typical smartphone cable instead of USB-B cable, this will bypass superspeed contacts and go regular usb port only contacts, did not work on my drive though, it spins but not recognized.

++ my attempt to clear relo-list with patch 2 in hddsuperclone has failed multiple attempts
preparing read
command failed

most likely because the controller is not SATA but the USB bridge cheapo
hddsuperclone is probably geared to SATA drives
usb mode not listed in the patch options ( only in the main cloner tool )
ati and scsi patch options tried those
encryption is not enabled on the drive
the script reads disk info and smart though
but bombs on that patch 2

so the possible solution is convert controller to SATA
or get SATA controller
and then try the patch 2 again
or use PC-3000 patch option ( which needs sata conversion as well )

what an effing mess.

I will never buy WD garbage again, no matter what...

So, I plotted myself a FREE solution for bringing that slow WD Passport back up to normal speed
- solder SATA connector to controller
- using hddsuperclone apply clear relo-list patch
- use ddsrescue to pull data off the drive
- rehab the drive

will post results on this WD adventure safari

3 wd drives: 2 out 3 so far success rate, third one in progress
after that, no more wd

That could make sense. And yes, I could get a $12 PCB board from Aliexpress, but is there a way to tell right away whether it's some memory/firmware module that's fried and not the BIOS chip which I would need to transplant anyway? Doing so would only reduce my chances as the risk of damaging what remains of the BIOS/PCB is non-negligible.



Good luck! I wish I was in your situation, but unfortunately I can't mount the drive so it doesn't like like a typical "slowness" issue.

what then? each brand has its typical weaknesses, just as cars or anything else... unfortunately :s
also, why would you clone it using ddrescue once you fired up hddsuperclone? The latter seems to be a more sophisticated thing.

_________________
Adatmentés - Data recovery

If you look at the amount of problems/questions per brand - WD is the outlier, although it maybe because they sell more drives. I got 4 of their drives and
all 3 with problems. The word is Seagate is better rated.

ddrescue I like because I can stop it and run it later where it left off, ie equivalent of pause button.
hddsuperclone I have not used before, need to look to see if it can be paused.

++ forgot to mention, how I know WD are crap, well I just remembered, WD MY BOOK, effing own WD power supply failed first, then cheap WD usb cable went out, and! I had the sucker stored in a effing plastic bag and handled with gloves... effing garbage.
Maybe now they factored in most of the problems and improved the design.

+++ took WD passport ULTRA out of it's enclosure, no shock absorbers inside, just cheap plastic... no sata interface, usb bridge, supposed to be a higher end drive - you drop that one time - heads out of alignment high probability.

Not the person you asked, but sharing the same feeling nonetheless: as a consumer, I buy a storage product so that I can trust that my data is safe and available, and that the manufacturer made its best for it to remain this way as long as the technology permits. In the case of this "WD My Passport", I absolutely feel cheated as a consumer. WD implemented not one but two mechanisms that clearly go against the above (and incidentally, my most essential interest and expectation as a consumer): the removal of a standard SATA connection, the implementation of a convoluted (and defective) firmware/PCB that encrypts my data without consent nor documentation.

Neither measure is justified from a cost, reliability or form-factor point of view, and represent flaws that were not stated on the packaging. It took the drive to be in an irrecoverable state for me to realize what was going on, so it's not even a case of failing to use my customer right to return a product bought by accident.

I've owned a fair amount of external hard drives from various manufacturers over the years, which, for all intent and purposes, were practically a standard SATA drive in a branded case. I couldn't imagine WD to be different, and in fact, I'm not aware of any manufacturer going to such lengths to obfuscate and lock users out from their own data. If you can explain to me what I'm missing and why this is all fine, I'm eager to be enlightened.

there are a couple of ways to tackle this. Will list two
1)image it and decrypt it with reallymine
2)find out sata points and convert the drive to sata. Image it and decrypt it.


to know whether you require decryption or not, just scan with any DR software. If you are seeing your data then its pretty straightforward extraction.

--