Forum rules
When asking for technical support:
- Search for posts on the same topic before posting a new question.
- Give clear, specific information in the title of your post.
- Include as many details as you can, MOST POSTS WILL GET ONLY ONE OR TWO ANSWERS.
- Post a follow up with a "Thank you" or "This worked!"
- When you learn something, use that knowledge to HELP ANOTHER USER LATER.
Before posting, please read https://www.cgsecurity.org/testdisk.pdf
I recently accidentally started clearing a 12TB USB HDD in Unraid. I stopped the process about 50GB in. Now the exFAT partition is lost.
I let TestDisk run in Windows but it was painfully slow (about one sector ever second making the whole process take ~16 days). I spun up TestDisk on a Ubuntu VM and it was a great deal faster. In one night it got to 28% and stopped there. What can I do?
Disk /dev/sda - 12 TB / 10 TiB - CHS 11444192 64 32
Analyse cylinder 3231422/11444191: 28%
check_FAT: Bad number of sectors per cluster
Linux filesys. data 2122984106 2124570474 1586368 [ M-!^SU, ~Zqz2SnR]
Unknown 2282654563 13995161851661154 13995159569006592 [x=M-b6hdT~IWt~KT
jM-&^BM-A~[7M-g,!M-&~Sn]
check_FAT: Bad jump in FAT partition
Unknown 3494497860 5900750147 2406252288
Mac HFS 4037987513 4071843816 33856304 [M-yM-( ~J^_M-F,,u~~@^?~?~~?^?~@^?~@^F~A^Aj^D,!M-
check_FAT: Bad number of sectors per cluster
check_FAT: Bad jump in FAT partition
Unknown 4537125218 13161230240097 13156693114880 [M-5_SYM-d]
Unknown 4848128477 6541814149 1693685673
check_FAT: Bad number of sectors per cluster
check_FAT: Bad jump in FAT partition
check_FAT: Bad number of sectors per cluster
Unknown 5769683592 144277601031 138507917440 [~YAM-,>M->&]
Unknown 5988127937 10743638609597632 10743632621469696 [K :2aM-~āh^ZM-%^RM-0M-nM-.HM-hM-k^
Xs ^^$L
Edit: I'm also running PhotoRec. No files found after one hour. When this is done in 22 hours I will try TestDisk another time. If that fails I might have to bite the bullet and pay 250+€ for data recovery.
DesertCookie wrote: ↑19 Jan 2021, 09:48
I recently accidentally started clearing a 12TB USB HDD in Unraid. I stopped the process about 50GB in. Now the exFAT partition is lost.
I let TestDisk run in Windows but it was painfully slow (about one sector ever second making the whole process take ~16 days).
Even with sectors as big as 4096 bytes you would rather end up with 5,6GB instead of 12.000GB after 16 days.
I spun up TestDisk on a Ubuntu VM and it was a great deal faster. In one night it got to 28% and stopped there. What can I do?
You can check the drive health. May you have to duplicate it because of its number of erros.
Even with sectors as big as 4096 bytes you would rather end up with 5,6GB instead of 12.000GB after 16 days.
It was the quick search. Looking at how long it took on Windows that's what I calculated. I am surprised Linux did it so much fast (W10: 1 sector/second, Ubuntu: 80-100 sectors/second).
You can check the drive health. May you have to duplicate it because of its number of erros.
The drive was manufactured in September of last year and has less than 200 hours of runtime. According to CrystalDiskInfo and Unraid it is as healthy as can be. I really think the only problem is of logical nature here.
PhotoRec seems to be more successful at scanning over the drive. It has recovered ~350 files after 24 hours and estimates it'll take 3 more days (instead of 24h total before). I'll let it recover whatever it can before running another TestDisk partition scan and hopefully recovery. I don't have the drive capacity to copy the entire drive in case the repair goes bad. I hope PhotoRec can find most of the 8TB on there which will fit on my NAS.
Even with sectors as big as 4096 bytes you would rather end up with 5,6GB instead of 12.000GB after 16 days.
It was the quick search. Looking at how long it took on Windows that's what I calculated. I am surprised Linux did it so much fast (W10: 1 sector/second, Ubuntu: 80-100 sectors/second).
You can check the drive health. May you have to duplicate it because of its number of erros.
The drive was manufactured in September of last year and has less than 200 hours of runtime. According to CrystalDiskInfo and Unraid it is as healthy as can be. I really think the only problem is of logical nature here.
I only accept trustable information like a smartmontools log file. What people think and tell does not always fit reality.
PhotoRec seems to be more successful at scanning over the drive. It has recovered ~350 files after 24 hours and estimates it'll take 3 more days (instead of 24h total before). I'll let it recover whatever it can before running another TestDisk partition scan and hopefully recovery. I don't have the drive capacity to copy the entire drive in case the repair goes bad.
I don't know what you mean by "the repair goes bad" and how copying "the entire drive" would solve the problem.
TestDisk recommends mirroring the drive. I don't have enough storage to do that. Instead, I wanted to recover what can be recovered (even though now I have 3TB of recovered data without any file names - 3TB sounds like most of the stuff I had on there). I'm trying to do that.
Currently, TestDisk stops at different percentages. I'll send the drive in for recovery, then RMA it.
Where do I go from here?: (last time I was at this point I changed D for deleted to something else and hit enter but TestDisk just displayed the main menu and nothing had changed).
Having searched for EFI GPT (on the drive itself):
Having searched for Intel (on a mirrored drive, in case the OG drive had bad sectors and TestDisk would abort):
DesertCookie wrote: ↑13 Feb 2021, 17:55
Where do I go from here?: (last time I was at this point I changed D for deleted to something else and hit enter but TestDisk just displayed the main menu and nothing had changed).
Having searched for EFI GPT (on the drive itself):
You have to use the p-key ("list files") to check each partition to see if it contains your content. If you don't find your content changing the status of the partition in question won't bring back your files.
Only the first partition starting at sector 2048 may be the right candidate. The other partitions do not fill up the disk space.
As your drive shows a sector size of 512 bytes, the Intel partition scheme would only support 2TB. Based on your statement having had just one partition (12TB) GPT musst have been used.
You have to use the p-key ("list files") to check each partition to see if it contains your content. If you don't find i
Thank you for your insight. I didn't realise I had to press P on the keyboard and instead had used the arrow keys to change to P which did something different.
Only one of the GPT partitions had data on it - none of it was what I was searching for. Is it possible TestDisk fails with partition recovery while PhotoRec still finds some files? Can I potentially change some drive geometry settings? I noticed sector size was 512 while I'm sure I selected the exFAT default when formatting which wuld be 4096.
