Distributed password cracker for operating over high latency networks of loosely coupled hosts. Main components used are django for the interface, celery for the communication and hashcat for the actual password cracking.

• P0rk will manage everything, run this on one server (doesn't need a GPU).
• Run Crackling on each of your GPU instances (one per server).

• rabbitmq-server
• your database of choice (postgresql, mysql or sqllite)

You can use pip install to download them

• django
• django-celery
• django-dajax
• database package, eg. psycopg2 (for postgresql)

Create new a new user and database for p0rk eg.

$ psql template1
> create user p0rk with password 'p0rkpassword';
> create database p0rkcrackling;
> grant all privileges on database p0rkcrackling to p0rk;

Remember to set database setings in the p0rk/settings.py file.

Tell p0rk to setup the database

$ python manage.py syncdb

Keep in mind that amqp running on rabbitmq should probably have a password set.

Currently there is no security authentication or authorization built into p0rk-crackling. I recommend that you run it only on trusted networks or within a VPN.

To run the web interface, just for testing or a low usage/no secruity server, run python manage.py runserver from within the django root folder.

To collect results you need to run a celery worker process, by default (can change in the settings.py file) it will run every seven seconds. python manage.py celery worker -B -l info -Q beats

• celery

• Hashcat (http://hashcat.net/hashcat/)
• oclHashcat-lite (http://hashcat.net/oclhashcat-lite/)
• Note: you need to make sure that you have the correct drivers for hashcat.

• Configure CRACKLING_BROKER, CRACKLING_BACKEND to point to the location of the rabbitmq amqp server.
• Configure HASHCAT_PATH, OCLLITE_PATH with the location of hashact and oclhashcat-lite respectively.

• Run hashcat and ocllite and accept the licence so that the eula.accepted file is created in the same folder as the hashcat binary.

Back one directory from the crackling folder (in p0rk-crackling/ if you just grabbed off github) run celery worker --app=crackling -l info -Q crackling