cgsecurity.org

Hey, I have somewhat of a similar problem with external disk: Seagate [ST4000DM000 / 1F2168 4TB] I am using for archiving the material. It was enclosed in external case, the connector got slowly loose and got tore down. I bought a docking station and connected it to the computer, resulting in one raw and one unallocated partition:

I searched around for the solution and came up to this advice: https://html5.litten.com/updated-how-to ... ecame-raw/.

When I ran Analyses in the Testdisk I got the following warning and error message, which I have overlooked the first time.


I have read on one forum this might be because of the sector size change:

This is acting like the drive was one of those external drives that when in the external unit the drive reports 4K physical / 4K logical sectors but when you pull it out and connect it directly to a SATA port it reports 4K physical / 512 logical and then the OS shows it as unformatted because of the sector size change.

I did the Quick search showing that partition exists:


And partition table is as it was when the disk was still bootable (including subfolders and files)


I wrote the partition table to the disk and restarted the computer (as in upper link). When reconnecting nothing changed. The disk is still not bootable, but now WIN10 shows one unallocated, one raw and another unallocated partition:


I repeated all the steps: partition table still shows the same folder and file order, but this time I am hesitating to write the partition table to the disk as I don't know what went wrong in the first place. Is there something else (besides warning/error messages) that I have overlooked? A friend suggested I should firstly make a backup of MBR? Or should I just copy all the files to the new disk?

Thanks to whomever helps; the disk is an archive of all my film projects from before 2010, so I really need a helping hand

Your disk was using 4096 bytes sectors. When using PC Intel partition, it's the only way to deal with disk over 2 TB.
To avoid the problem, it's recommanded to use EFI GPT.
Run TestDisk, select your external disk, choose EFI GPT, Analyse, Quick Search, if you see your partition, choose Write, confirm, Quit and restart your computer.

Hey admin, I did as told and now the disk keeps running and Testdisk stays idle/stuck at the following cylinder:

Disk /dev/sdd - 4000 GB / 3726 GiB - CHS 486401 255 63
Analyse cylinder 486401/486400: 99%

Warning: number of bytes per sector mismatches 4096 (NTFS) != 512 (HD)
Warning: number of bytes per sector mismatches 4096 (NTFS) != 512 (HD)
MS Data 16384 7814021119 7814004736 [FILM PROJECTS]

After waiting several hours, I tried closing Testdisk, was unresponsive; same happened in Task manager, also the system didnt want to shutdown, needed to wait the disk stopped being active before restarting the computer.

Dear Chris, thanks a lot for your help. I have succesfully recovered the EFI GPT partition table in Ubuntu OS and managed to copy all the (lost) files to the new disk - for backup. I also hapily donated to the cgisecurity

I have got one last question however, that confuses me. When I tried to open the disk and check the files in WIN10 the disk mounted - disk management reported the partition as a whole but in RAW format. I ran TestDisk again to try to write the EFI GPT partition table also in Windows, but got to the same cylinder error. Considering all my files are now safely stored on a backup drive, should I just ignore the recurring bug and format the disk in NTFS to use it as a fresh disk or is there some other way to trifle around the problem?

Any suggestions for a program that would wipe the disk clean, possibly with only zeros before reformatting it back to NTFS?

Thanks for your help,

best
D

Use the linux dd command to copy an endless stream of zeros to your drive using the zero device /dev/zero.