Hashtopolis Forum
Issue with bruteforce mode and static chunking - Printable Version

+- Hashtopolis Forum (https://hashtopolis.org)
+-- Forum: Support (https://hashtopolis.org/forum-1.html)
+--- Forum: Problems (https://hashtopolis.org/forum-2.html)
+--- Thread: Issue with bruteforce mode and static chunking (/thread-34.html)



Issue with bruteforce mode and static chunking - Nemo - 11-06-2018

Hello.
There is some information about my Hashtopolis configuration:
- Server and agents are running on Ubuntu 16.04 LTS;
- All agents have Nvidia GTX 970;
- Hashtopolis v0.8.0 (C# agents);
- Hashcat v4.2.1;
- CMDs like: -a 3 #HL# -1 ?l?u?d ?1?1?1?1?1?1.

So the problem is when I try to run the task with type of cmd as mentioned above and the static chunking (using static number of chunks, e.g. 8), the task will have 100% dispatched keyspace, every chunk will be assigned with 1 agent, but only 12.5% (or multiple of 12.5 like 25, 37.5 etc: N * 100 / 8 < 100) of dispatched keyspace will be searched, so the hash couldn't be found. But if the hash of correct password is in the range of first dispatched and searched keyspace, password will be cracked. IDK if this problem is only with bruteforce mode or not, but it takes place for different types of hashes (like RAR5, PDF, HCCAPX, Office etc).

If there is no chunking with static chunk size, the result is unpredictable, but in most cases is correct. 


P.S. There is an another small problem: when creating hashlist of HCCAPX type there is an incorrect display of hccapx info in the web interface. It should be like [keymic*mac_ap*mac_sta*essid], but it's only [mac_ap*mac_sta*essid] without keymic.

Thank you.


RE: Issue with bruteforce mode and static chunking - s3in!c - 11-06-2018

The first problem is that you are using the C# agent, because this one is not really up-to-date and there might be a lot of problems being caused by this one.
You definitely should only use the python one at the moment because it's developed in parallel with the server.

I'm not sure if I understand your problem correctly, but when you create 8 chunks, the checkpoint normally is just at the end of the chunk which then normally would be 12.5% and it's multiple. Also I don't understand what you exactly mean with 'the result is unpredictable'.

I think the best would be that you try again what you wanted to achieve with the python agent and see if you still experience the same issues and then report back here if you still have trouble. Also it would help to know what kind of hashes you tried to attack and if there is some strange behavior of the agent, you could also provide some debug logging output (add the debug flag in config.json)

I will check the thing with the hccapx.