🚀 go-pugleaf

In comp.security.unix Raphael <rraphael_pierre@hotmail.com> wrote:
> It was stated that the inclusion of the salt in the UNIX password scheme
> increases the difficulty of guessing by a factor of 4096.  But the salt
> may be stored in plaintext in the same entry as the corresponding
> ciphertext password. Therefore, those two characters may be known to the
> attacker and need not be guessed.  Why then is it asserted that the salt
> increases security?

The two salt characters are prepended to the password prior to
encryption. This has a couple advantages:

1. It discorages building a code book by requaring you to
store 4096 entries for each possibible password.

2. It also prevents identical passwords from generating identical
entries in /etc/password. For each guess, you need to do a crypt(3)
for every salt which appears in the password file. crypt(3) being the
most time consuming part of the process by far, salt really slows
things down.

There are a couple other minor bonuses from salt, but no show
stoppers. In short, it makes breaking arbitrary passwords harder. For
brute forcing a single entry, it's "only" as hard as breaking DES. 

-- 
Matthew Gauthier <elleron@crosswinds.net>

Raphael <rraphael_pierre@hotmail.com> writes:
>I have a question for a assignment that is due tomorrow morning that I
>can't figure out.  I'd appreciate any help.
>Here goes:

>It was stated that the inclusion of the salt in the UNIX password scheme
>increases the difficulty of guessing by a factor of 4096.  But the salt
>may be stored in plaintext in the same entry as the corresponding
>ciphertext password. Therefore, those two characters may be known to the
>attacker and need not be guessed.  Why then is it asserted that the salt
>increases security?

>I have a feeling this is easy but I can't figure it out.  Can anyone
>help?

The salt protects against a systamatic dictionary attack against a
long list of passwords. It ensures that if two users both pick the
word 'elephant' for their passwords, that the crypted password entries
for both of them won't be identical (if the random generator is good
enough, there's only a 1:4096 chance in them being identical).

Instead, the password hacker has to scan through the list of crypt'd
passwords, find the salt for each one, and then seed the crypt
algorithm with the salt for each password entry and try all the
dictionary words. This is instead of crypting each dictionary entry,
and then comparing it to every password in the list. 


--
Doug McIntyre						merlyn@visi.com
  Network Engineer/Tech Support/Jack of All Trades of Vector Internet
Due to circumstances beyond your control, you are master of your fate
and captain of your soul.

On Mon, 13 Dec 1999 09:14:08 GMT, Raphael <rraphael_pierre@hotmail.com> wrote:
>I have a question for a assignment that is due tomorrow morning that I
>can't figure out.  I'd appreciate any help.

I'll try to give a hint...

>It was stated that the inclusion of the salt in the UNIX password scheme
>increases the difficulty of guessing by a factor of 4096.  But the salt
>may be stored in plaintext in the same entry as the corresponding
>ciphertext password. Therefore, those two characters may be known to the
>attacker and need not be guessed.  Why then is it asserted that the salt
>increases security?
>
>I have a feeling this is easy but I can't figure it out.  Can anyone
>help?

It depends on the nature of the attack being considered.  For a
single software-based attack on a single password the salt
doesn't help, but consider either an attack by DES hardware
(which is generally faster than software implementations, but
only computes the ".." salt), or a dictionary attack where one
pre-encrypts a "word" list...

		--Ken Pizzini