Hacker News new | past | comments | ask | show | jobs | submit login

The assumption is that you have access to the hashed version of the password. That's not so hard to get from some sites. I have retrieved user-password pairs using very simple SQL Injection in some e-commerce sites(not any the big ones, of course).

The numbers give make no sense, because doesn't state which hash is using, and the difference may be huge:

~/john-1.7.2/run$ ./john --test Benchmarking: Traditional DES [128/128 BS SSE2]... DONE Many salts: 906828 c/s real, 908646 c/s virtual Only one salt: 805504 c/s real, 805504 c/s virtual

Benchmarking: BSDI DES (x725) [128/128 BS SSE2]... DONE Many salts: 31271 c/s real, 31334 c/s virtual Only one salt: 30617 c/s real, 30617 c/s virtual

Benchmarking: FreeBSD MD5 [32/32]... DONE Raw: 8617 c/s real, 8652 c/s virtual

Benchmarking: OpenBSD Blowfish (x32) [32/32]... DONE Raw: 415 c/s real, 416 c/s virtual

Benchmarking: Kerberos AFS DES [48/64 4K MMX]... DONE Short: 186368 c/s real, 186741 c/s virtual Long: 528588 c/s real, 531779 c/s virtual

Benchmarking: NT LM DES [128/128 BS SSE2]... DONE Raw: 6575K c/s real, 6588K c/s virtual

Which I find unbelieveable is that lots of web applications use simple MD5-passwords(not the FreeBSD MD5 based version but just MD5 hashes) without even using salts, which makes them almost instantly crackable using Rainbow tables.




If you can get access to the hashed password, that's a major breach in of itself.

It's almost comparable in my mind to saying, if you get physical access to the machine you can do exploits X, Y, and Z. Well, yeah, if you can get that far, you've pretty much won.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: