AES password cracker

[npoe1]

I have some old files from when I was in college and in a very clever movement I encrypted the files using AES256 just for the "fun". I forgot the last part of the password and the combinations that I have tried haven't worked.

Does anyone knows if there is somekind of password cracker for AES? I have found MD5, SHA1, etc. but not for AES.

Thanks,

 

[Gooberlx2]

Truecrypt container or something?

If you know most of the password, then your best bet is going to be some sort of brute forcing tool on the remaining parts.

How much of the password did you forget, and how complicated did you allow it to be? symbols, special characters, etc... If you forgot more than a few characters in a complicated string, it just might take longer to break through than your natural life.

 

[npoe1]

It was is 14 or 15 characters length password. I'm sure of the first 10 characters but the remaining 4 or 5 are what I can't remember. I have an idea of what it could be and a force brute attack could work in the remaining characters but I can't find a tool for that to AES encrypted files.

The program I used encrypts the file without modifying the extension and I have seen some for RAR and Zip but they seem to take in account the file format.

 

[Gooberlx2]

I guess it depends on what tool you used. If you still have that tool, and it can run via command line, then you could probably write a script of some kind to get the job done. If you don't still have the tool or an alternative for it, you're probably out of luck. AES encryption is open and standard, but a program's implementation isn't necessarily so.

If it's literally just the file contents encrypted (no proprietary container format of some kind), then there may be a variety to tools to do so. It can still depend on the exact cipher used aes-256-cbc vs aes-256-ecb, etc.... It would be best if you can remember how exactly you encrypted it in the first place.

Still, according to here, 5 complex characters could take just shy of 3 years to break at 100 passwords per second. Multiply that by each cipher you have to try.

 

[Mark R]

The difficulty in finding a cracker will be knowing how your encryption program derives the encryption key from the password.

Some low-end programs use something like SHA-256. Better programs use a salted multi-iteration process like PBKDF2. The reason PBKDF2 is used is because it's super slow, so it can't be optimised and automated brute-forcing will just crawl along.

To crack it, you
1) really need to know the exact key derivation algorithm including salt, so as to produce optimised code
2) Knowledge of what was in the file to start with (i.e. known plaintext)

With a bad key derivation function, like SHA, a cracker could test millions of keys per second. This will give you a good chance of cracking 5 "good" characters within a few hours.

If your program used PBKDF or something similar, and your cracker can only test a few thousand passwords per second - you'll be waiting weeks or months for 5 unknown characters.

If you are stuck scripting your program which tries to decrypt the file, and it attempts decryption and then checks a CRC - so that you can test 10 passwords a second - well, you're screwed.

 

[npoe1]

Looks like I have a lot to if I really want those projects back. Hopefully I'm going to find out the details you mention during the day.

 

[ringtail]

A tad expensive at $ 300, but still a solution to the problem:
Elcomsoft Forensic Disk Decryptor (http://www.elcomsoft.com/efdd.html)

 

[Mark R]

Nah. This won't help.

Crackers need to be designed for the exact file format being cracked. The one linked only does encrypted partitions. More importantly it only accesses encrypted partitions when the encryption keys are known and available.

In this case, the op has lost the keys, so this approach won't work.