|
Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com |
Re: Re[2]: password aging
Aleph One (aleph1
dfw.net)
Wed, 2 Sep 1998 10:38:10 -0500 (CDT)
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
- Next message: Firewalls: "New one: Securing an HTTP server"
- Previous message: Adam Shostack: "Re: Reverse Proxying of FTP?"
- In reply to: Firewalls: "Reverse Proxying of FTP?"
- Next in thread: Ryan Russell: "Re: Re[2]: password aging"
On Wed, 2 Sep 1998 Steve.Bleazard
wdr.com wrote:
> One alternative to password aging, is to force everyone to use a
> password generator. FIPS181 from the US government describes (and
> implements) such a generator. I have found the FIPS181 algorithm
> generates good pronouncable passwords. They are also far less
> susceptible to social engineering.
>
> Using password generators has many problems in itself, not least of
> which is the tendency for people to write the password down. However,
> if security demands good password aging and system wide password
> re-use detection, then the local policies can be enforced to deal with
> this and a generator is a viable alternative.
This reminds me of this little blurb that comes with the Crack programs.
From doc/fips181.txt:
Federal Information Processing Standard 181 defines a standard for an
automated password generator to be used in "all federal departments
and agencies where there is a requirement for computer generated
pronouncable passwords"... for passwords of between 5 and 8 characters
long.
Basically it's a generator which takes a good PRNG and a bunch of
fixed syllables (composed from lowercase ascii letters) and uses the
former to drive concatenation of the latter, producing at the business
end a "pronouncable password".
Reading FIPS181 (http://csrc.ncsl.nist.gov/fips/fips181.txt) one gets
a good feel for the reduction in search space that this algorithm
provides to the password cracker.
Section 2.4 cites that the algorithm is capable of producing
"approximately 18 million 6-character" passwords; compare this with
the set of 309 million lowercase 6-character passwords, and we see
that the lack of entropy in the output has reduced the search space to
about 5% of it's original size.
Interesting; from this basis we may pose the following student project:
or values of N constrained by
your resources.
3) sort/uniq, dawg and gzip this dictionary and put it up on an
Internet FTP site, posting an announcement of a new Crack dictionary
containing all possible N-character plaintext federal passwords.
4) Write an essay describing your experiences of consequent federal
investigation, backbiting and paranoia.
--
To verify the feasibility of (3), the author can confirm that the
highly redundant 2Gb dictionary of all possible 6-character lowercase
passwords (newline separated) compresses to about 7Mb under dawg/gzip.
YMMV.
As you can see using FIPS181 is a very bad idea.
> Steve
Aleph One / aleph1
dfw.net
http://underground.org/
KeyID 1024/948FD6B5
Fingerprint EE C9 E8 AA CB AF 09 61 8C 39 EA 47 A8 6A B8 01
- Next message: Firewalls: "New one: Securing an HTTP server"
- Previous message: Adam Shostack: "Re: Reverse Proxying of FTP?"
- In reply to: Firewalls: "Reverse Proxying of FTP?"
- Next in thread: Ryan Russell: "Re: Re[2]: password aging"
This archive was generated by hypermail 2.0b3 on Sat Jul 17 1999 - 07:11:46 CDT