Need to save a Hard Drive Controller not the HDA

[DontBeThatGuy]

I should add that HDDHackr overwrites sectors 16-22 with the HDDSS.bin contents, so you would need to apply the hack before cloning your drive.

Thanks for the heads up. I wasnt sure if i should clone it before or after applying the HDDSS.bin patch. Will let you and anyone else reading know how it goes soon...

[DontBeThatGuy]

Well, Changing the Brand/Model/Serial# worked but at some point the software must check the firmware or run a crc check on it somehow. But it did get past one check atleast.

[fzabkar]

Well, Changing the Brand/Model/Serial# worked but at some point the software must check the firmware or run a crc check on it somehow. But it did get past one check atleast.

Does that mean that when the hard drive fails, you end up with an expensive pile of junk? :x

[DontBeThatGuy]

Well, Changing the Brand/Model/Serial# worked but at some point the software must check the firmware or run a crc check on it somehow. But it did get past one check atleast.

Does that mean that when the hard drive fails, you end up with an expensive pile of junk?

Pretty much. Theres already 4 bad sectors.

[DontBeThatGuy]

What if i take this spare Toshiba MK3276GSX and remove the rom chip and edit in HxD the serial # to match the Original im trying to clone. The firmware version is the same but the rom chip is a winbond and the original is an ST. I think it worst case senerio i would just have to reprogram it if it has some sort of crc. What do you guys think?

[DontBeThatGuy]

Original is 4mb ST Micro 25P40VG
Spare is 4mb Winbond 25X40BL

[fzabkar]

What if i take this spare Toshiba MK3276GSX and remove the rom chip and edit in HxD the serial # to match the Original im trying to clone. The firmware version is the same but the rom chip is a winbond and the original is an ST. I think it worst case senerio i would just have to reprogram it if it has some sort of crc. What do you guys think?

I think it would be advisable to backup the ROM, given the circumstances. That is, since you can't backup your drive, you may as well prepare yourself for a data recovery should the need arise. You must not damage the ROM, though.

I've had a look at some Toshiba resource dumps and ISTM that the ROM is comprised of several configuration pages (CPs ?). Each CP appears to have its own 8-bit CRC byte. To assist us in understanding the CRC, you could enable PUIS (Power Up In Standby) and examine the ROM contents before and after. The difference should be in the PUIS flag and a corresponding CRC byte.

There appear to be two serial numbers. One is reported via an ATA Identify Device command (CP56) whereas the other (CP9A) is probably only accessible via a Vendor Specific Command (VSC). As Spildit has suggested, it seems unlikely, but not impossible, that a VSC would be used by the OEM.

If the bad sectors are "fake", then you may be able to read them with an ATA READ LONG or SCT READ LONG command.

[DontBeThatGuy]

Yes I used Acronis TI 2014 rescue cd to do a sector by sector clone.
There are a few partitions that are Bitlocked so sector by sector is a must.
I obtained the Bitlocker unlock keys so i know the cloned data must be good or i dont think they would unlock.
During the clone process there were 4 bad sectors Acronis ignored. location was
214,872,205
221,078,858
221,078,859
221,078,860

Can you please explain the reason you do not think dumping the rom chip and editing the serial number will work? I am not talking about swapping the rom chips, just changing the serial number on the spare drive to match the original. I do not understand the limitations to know why this could not work.

I started my post before fzabkar replied. Thanks for that detailed info.

[DontBeThatGuy]

Spildit,

Changing the serial, fw version, and make/model that fzbarkar helped me with (thanks fzabkar) Got me past the initial error. But right toward the end of the software loading it hangs at a specific screen. After awhile it eventually does a shutdown and upon a reboot, it does some strange thing where a loading screen is just an empty background screen. Whats even more strange is I am able to get into explorer and all the partitions are unlocked. Whats even more strange is if i use a different hard drive the loading screen is displayed and loads up to about 99% but throws a critical error at about 10% but contines to load. But if i trigger explorer the partitions are locked.

I think the hddss.bin fzabkar edited got me past the first check but i noticed in HxD the edited hdss.bin has the serial and firmware version joined together where the original looks to have bytes inbetween them. I dont know if that was intentional or not.

[fzabkar]

Some parameters are left justified, others are right justified. That's up to the HDD manufacturer. I merely transferred the appropriate fields from CrystalDiskInfo's Identify Device output.

See the ATA command standard for a detailed explanation of the structure.

http://www.t13.org/documents/UploadedDo ... A8-AST.pdf

[DontBeThatGuy]

If they are "fake bad sectors" would i need to unlock or decrypt the partitions to determine this? Reason i ask is because the software does some sort of automated unlocking and if i was to decrypt the partitions it might mess up the automated unlocking in the program and might not work anymore. If there is a way to unlock and clone on the fly i might consider trying that but i havent found too much info on doing this without decrypting which i dont want to attempt.

[fzabkar]

Bit #1 of Identify Device word 206 (0x003D) is indicating that the Toshiba drive does not support SCT Read/Write Long (Long Sector Access). The WD drive does, though (0x303F). Perhaps READ LONG still works. It would be worth trying both in any case.

If you're intent on safeguarding your investment, you might like to consider engaging someone to backup the drive's firmware resources, although I think the ROM contains the majority of the critical data. Backing up the ROM is something I would definitely do.

[DontBeThatGuy]

I finally got a hold of a company that can get old stock of the drives. I was quoted a few hundred. Problem is, i know some people who got replacement drives while still under warranty and they were sent refurbished drives. If im going to pay a few hundred i would prefer brand new. So i dont know if they only send out refurbished when replaceing under warranty, or if they were cutting costs.

I really would prefer to figure this out and maybe even get this cloned onto a solid state drive if at all possible. Im starting to think thats not possible at this point.

[fzabkar]

I really would prefer to figure this out and maybe even get this cloned onto a solid state drive if at all possible. Im starting to think thats not possible at this point.

I have one more idea, but I would need to see the Toshiba ROM and the UNDO.BIN file produced by HDDHackr for your WD drive.

[DontBeThatGuy]

Ok I dont know if fzabkar's edited hddss.bin had to read like this or not but i think this may be the culprit.

If you look at the original here: http://pastebin.com/raw.php?i=eB6vGKeF
it reads ..@...GS001A

And if you look at the WD after the hddss.bin has been applied here: http://pastebin.com/raw.php?i=Sf5u47vS
it reads ..@..2GS001A

I am thinking this might be the problem but i could be wrong?

[DontBeThatGuy]

I really would prefer to figure this out and maybe even get this cloned onto a solid state drive if at all possible. Im starting to think thats not possible at this point.

I have one more idea, but I would need to see the Toshiba ROM and the UNDO.BIN file produced by HDDHackr for your WD drive.

Would the ROM from the same exact make/model/firmware version suffice? I am alittle reluctant to desolder the chip from the original until i know for sure the method i choose is safe. I have a willem and an arduino uno i could use to build a device to interface with the drive/chip.

[fzabkar]

Word 22 of the Identify Device information block was defined in early versions of the ATA standard as "Number of vendor specific bytes available on READ/WRITE LONG cmds". In current versions it has been rendered obsolete.

In your case it would appear that the Toshiba drive does not support the READ/WRITE LONG commands, but it would be worth trying anyway.

[fzabkar]

Would the ROM from the same exact make/model/firmware version suffice? I am alittle reluctant to desolder the chip from the original until i know for sure the method i choose is safe.

Let's just work with the UNDO.BIN for the moment.

[DontBeThatGuy]

Would the ROM from the same exact make/model/firmware version suffice? I am alittle reluctant to desolder the chip from the original until i know for sure the method i choose is safe.

Let's just work with the UNDO.BIN for the moment.

PM sent.

[DontBeThatGuy]

Spildit WDR doesnt work with win 8.1 so i will have to install XP onto a spare computer. I will try and get that done in an hour or few.