hashcat Forum
LUKS benchmarks and estimated times - Printable Version

+- hashcat Forum (https://hashcat.net/forum)
+-- Forum: Support (https://hashcat.net/forum/forum-3.html)
+--- Forum: hashcat (https://hashcat.net/forum/forum-45.html)
+--- Thread: LUKS benchmarks and estimated times (/thread-7051.html)



LUKS benchmarks and estimated times - danr - 11-27-2017

I've been reading the docs, example benchmarks, and running hashcat locally against the LUKS header of my usb drive that contains some valuable data. The password is at least 10 characters, so I know this is a challenge. I'm trying to figure out if it is worth it to try and crack it, or if I should let it go. I've investigated cloud options like Google- and it seems like the Tesla GPUs they offer aren't great for cracking. The benchmarks I've seen posted for various GPUs don't include LUKS (mode 14600 doesn't seem to be in the default benchmarking list)

I'm pretty sure the password doesn't contain any real words, so narrowing it down with rules, etc will be hard.

It's hard to figure out how long cracking this thing might take on any given GPU/set of GPUs
Here's the output of cryptSetup:

Code:
sudo cryptsetup luksDump /dev/sdb2
LUKS header information for /dev/sdb2

Version:        1
Cipher name:    aes
Cipher mode:    cbc-essiv:sha256
Hash spec:      sha1
Payload offset: 4096
MK bits:        256
MK digest:      bb 49 14 91 26 e1 be 4e 45 2c 9e 81 15 95 45 43 14 1d 9c eb 
MK salt:        09 b3 d0 c4 15 8e cb 0b 4c 20 02 39 a3 71 7c 67 
                61 5c 3a ef 8b 3f f9 87 fb d5 bc 03 b9 eb ca 21 
MK iterations: 18750
UUID:          4f85fbe5-2d73-47e4-a59f-3ae3b080d913

Key Slot 0: ENABLED
Iterations:          75125
Salt:                3d cd 01 61 14 4b 33 a4 70 b7 a7 ca 5d ee 53 54 
                      db 82 f9 69 df cd 21 42 7f 6f 56 20 1f 45 9e cc 
Key material offset: 8
AF stripes:            4000
Key Slot 1: DISABLED
Key Slot 2: DISABLED
Key Slot 3: DISABLED
Key Slot 4: DISABLED
Key Slot 5: DISABLED
Key Slot 6: DISABLED
Key Slot 7: DISABLED

I'm willing to drop some cash on a cracking rig, or pay someone to use theirs, but I'm having a hard time figuring out how crackable this thing is. Running it locally on my macbook pro seems to suggest it will take many thousands of years, which isn't promising :[


RE: LUKS benchmarks and estimated times - royce - 11-27-2017

Mode 14600 (LUKS) benches at around 10k H/s on a GTX 1080 with -w 4 and -O.

If you have literally no idea what those 10 characters might be ... it ain't happenin' any time soon.


RE: LUKS benchmarks and estimated times - asirena - 11-27-2017

It will take many thousands of years.

According to the speed indicated by royce, using one of the most powerful graphics cards of the moment:

* if it was only digits, it'd take: 10¹⁰/(10KH/s * 60 * 60 * 24) = 11.57 days

* if it was only hex chars, it'd take: 16¹⁰/(10KH/s * 60 * 60 * 24 * 365) = 3.48 years

* if it was only lowercase letters OR only capital letters, it'd take: 26¹⁰/(10KH/s * 60 * 60 * 24 * 365) = 447.63 years

* if it was digits AND only lowercase letters OR only capital letters, it'd take: 36¹⁰/(10KH/s * 60 *60 *24 * 365) = 11,593.6 years

* if it was lowercase letters AND capital letters, it'd take: 52¹⁰/(10KH/s * 60 *60 *24 * 365) = 458,381.2 years

* if it was digits AND lowercase AND capital letters, it'd take: 62¹⁰/(10KH/s * 60 *60 *24 * 365) = 2,661,400.8 years

* if it was any char, it'd take: 95¹⁰/(10KH/s * 60 *60 *24 * 365) = 189,858,238 years

So you should not waste effort and / or money