Why crack LM hashes with the purpose of detecting and eliminating “weak” passwords at all, then?
Solar Designer: One possible reason would be to ensure that the passwords are not as weak as to be easy to guess from the remote (or from a local login prompt, for that matter). Another reason would be to ensure that the not-so-weak NTLM hashes of the same passwords (that Windows systems use along with or instead of LM hashes) would be strong enough to withstand certain offline attacks. Someone who is into Windows security (I am not) could explain those subtle reasons better. What is the situation in the Unix world? Solar Designer: Speaking of Unix passwords, even the ancient traditional crypt(3) hashes can withstand John runs on a single CPU if extremely complicated passwords are chosen. Although it would be taking over 100 years to exhaustively search the printable US-ASCII keyspace against traditional crypt(3) hashes on a single CPU (that is currently available), in practice it is common to crack 20% to 60% of such hashes within a reasonable time (days, weeks, or months). When attac
