Posts: 9
Threads: 3
Joined: May 2022
05-14-2022, 04:37 PM
(This post was last modified: 05-14-2022, 05:38 PM by phaphip.)
there is a wordlist containing 500-character long passwords.
it seems that
./hashcat.bin -m 0 hashes.txt wordlist
produces no result, even with a long password whose md5 hash is really in hashes.txt
how to work with long passwords and md5 (-m 0)?
Posts: 63
Threads: 1
Joined: Mar 2017
With hashcat? You don't. Hashcat kernels are largely constrained by what you can efficiently compute on a GPU, and the plaintext limit on the "pure" MD5 kernel is 256 bytes (the optimized kernel is 31 bytes).
Google "mdxfind" for a possible solution to your problem.
(05-14-2022, 04:37 PM)phaphip Wrote: i have a wordlist containing possible 500-character long passwords.
yes it is really needed for this challenge.
it seems that
generatelongpwds | ./hashcat.bin -m 0 hashes.txt
produces no result, even if i test with a long password whose md5 hash is really in hashes.txt
how to work with long passwords and md5 (-m 0)?
Posts: 9
Threads: 3
Joined: May 2022
05-14-2022, 05:36 PM
(This post was last modified: 05-14-2022, 05:40 PM by phaphip.)
(05-14-2022, 05:23 PM)pdo Wrote: With hashcat? You don't. Hashcat kernels are largely constrained by what you can efficiently compute on a GPU, and the plaintext limit on the "pure" MD5 kernel is 256 bytes (the optimized kernel is 31 bytes).
is there a way to recompile hashcat with this constant 256 replaced by 1024?
it is ok if it is slower.
or is 256 a real gpu hardware limitation?
would hashcat work with:
md5(known_fixedpart_len300 + pwd_len_max_256)
the fixed part of length 300 is always the same, and the generated passwords are of length less than 256.
at the end it is always md5(more than 256 char), i dont know if the fixed part changes something.