cudadbcracker - cracking using CUDA

cudadbcracker is a salted SHA-1 cracker for cracking database password hashes using CUDA enabled videocards. It was developed using NVIDIA CUDA Software Development Kit (CUDA SDK) Version 2.1.

cudadbcracker is free software under GNU GPL v3, currently supports MSSQL and Oracle 11g hashes.

What’s CUDA?

CUDA == Compute Unified Device Architecture

CUDA (an acronym for Compute Unified Device Architecture) is a parallel computing architecture developed by NVIDIA. CUDA is the computing engine in NVIDIA graphics processing units or GPUs that is accessible to software developers through industry standard programming languages.

More information: here (NVIDIA) or here (Wikipedia)

Features:

  • cracking multiple password hashes (max. 999)
  • session handling (cracking can be interrupted and continued later)
  • logging cracked password
  • optional character set selection

Quick start

  • Show command line options: cudadbcracker.exe
  • Show available GPUs: cudadbcracker.exe -v
  • Crack MSSQL hashes: cudadbcracker.exe -h mssql -c un -i mssqltest.hashes
  • Cracking Oracle 11g hashes: cudadbcracker.exe -h ora11g -c uln -k 2 -n 2 -i ora11gtest.hashes
  • Continuing a saved session: cudadbcracker.exe -l default.ses
  • Running tests: Start mssql_test.cmd or ora11g_test.cmd.

Command line options

------
cudadbcracker.exe -h <hashtype> -i <inputfile> -s <sessionfile> -l <loadsession> -r <record>
-c <charset> -n <numstreams> -k <kernels> -v -d <device>
------
-h(*), --hashtype=<hashtype>
        Type of hashes to load (mssql,ora11g).

-i(*), --inputfile=<inputfile>
        Name of input file.

-s, --sessionfile=<sessionfile>
        Name of session file. Default is "default.ses".

-l, --loadsession=<loadsession>
        Name of session file to load.

-r, --record=<record>
        Name of record file. Default is "default.rec".

-c, --charset=<charset>
        Character set used for cracking (can be any combination of 'u'pper, 'l'ower,'n'umeric and 's'pecial). Default is "un".

-n, --numstreams=<numstreams>
        Number of streams. Default value:1

-k, --kernels=<kernels>
        Number of kernels per streams. Default value:1

-v, --view
        View GPU information about CUDA enabled devices.

-d, --device=<device>
        Selected CUDA device index. If not given, the default is auto select.

------------------------
* = required parameter
------------------------

Note: When loading from session file, all other parameters can be skipped.

More information: here

Thank you Marcell Major - Project Leader of cudadbcracker -, for sharing this tool with us.