+- hashcat Forum (https://hashcat.net/forum)
+-- Forum: Deprecated; Ancient Versions (https://hashcat.net/forum/forum-46.html)
+--- Forum: Feature Requests (https://hashcat.net/forum/forum-7.html)
+--- Thread: Distributed oclhashcat version (/thread-327.html)
Distributed oclhashcat version - tatgdi - 03-08-2011
To take on the challenge of brute-forcing digest keyspace above 11 charachters via bruteforce method.
This would be possible if there was a distrubted version of oclhashcat, one could harness the power of sevral systems instead of 1 single box with ie x4 ATI 5970's(8 cores)/6990's(8cores) or even the upcomming Nvidia 590 GTX. Having multiple systems of the above calibre would be very resourcesful and pratical for bruteforcing.
RE: Distributed oclhashcat version - atom - 03-08-2011
this is already possible. see blandys successfull approach to get the idea behind: http://www.md5decrypter.co.uk/projects.aspx
RE: Distributed oclhashcat version - tatgdi - 03-10-2011
(03-08-2011, 06:04 PM)atom Wrote: this is already possible. see blandys successfull approach to get the idea behind: http://www.md5decrypter.co.uk/projects.aspx
Hey Atom,
I imagine the challenge would be in scripting the front end application that would allow oclhascat to become distributed; is this correct? If I am not mistaken have you not made hashcat into a distributed application? The following link that you provided is definitely a great start but what about locating a open-source front end that exists that would allow oclhascat to become distributed as an alternative option?
RE: Distributed oclhashcat version - atom - 03-11-2011
yes a lot people already asked for such an automatic solution and as you can see its absolutly possible. but i am focusing hashcat development. what i want to say is this can be done by some 3rd person.
RE: Distributed oclhashcat version - thorsheim - 05-16-2012
Elcomsoft and Passware does distributed, with a server that can be passive in that it only distributes tasks and collects 'replies', or it can also do cracking itself as an active node.
A 'server' would take as input the command line for -plus (or -lite), and break the total task into sub-tasks (sub-keyspaces) with an estimate of perhaps a couple of (gpu) minutes to finish each sub-task (depending on gpu on 'agent'). Results (hash:plain
alt:hex) is returned to server. For next sub-tasks to be handed out, found hashes will be reported/removed, so that each agent does not try to recover it anymore.or something like that. Morning, off to work. :-)
RE: Distributed oclhashcat version - atom - 05-16-2012
There are already some nice 3rd party distribution solutions for oclHashcat-lite for SL3. With some small changes, they could support all the other algorithms as well. All it takes is their will to rewrite it. Its also a very good example why it can be clever to source this thing out to some 3rd party.
See here:
http://www.servmg.com/uploaded/sl3cloud/sl3cloud4.png
http://winiu.com/support/sl3client/shot_000002.png
http://www.sl3net.com/cpanel/SL3NET-CPANEL-01.jpg
http://www.sl3net.com/cpanel/SL3NET-CPANEL-04.jpg
There are even more solutions, but their servers are down so I was unable to fetch more screenshots.
RE: Distributed oclhashcat version - KT819GM - 05-16-2012
@atom
Don't forget that they ask money for that like on sample pic's Martech ~20EU / pc.
----------------
If you earn for living by doing hash cracking ask them to support other algo's, people under that name have some nice brain power
There is some more sw (free, opensource) which can control these processes, just need to look at them and build something suitable.
RE: Distributed oclhashcat version - atom - 09-07-2012
Done with oclHashcat-plus v0.09: https://hashcat.net/forum/thread-1541.html
Thread closed