home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!ames!elroy.jpl.nasa.gov!usc!news!netlabs!lwall
- From: lwall@netlabs.com (Larry Wall)
- Newsgroups: comp.unix.admin
- Subject: Re: Password Aging
- Message-ID: <1992Sep3.163710.11783@netlabs.com>
- Date: 3 Sep 92 16:37:10 GMT
- References: <1992Aug26.163117.28008@iplmail.orl.mmc.com> <1992Sep2.124604.6904@mlb.semi.harris.com>
- Sender: news@netlabs.com
- Organization: NetLabs, Inc.
- Lines: 31
- Nntp-Posting-Host: scalpel.netlabs.com
-
- In article <1992Sep2.124604.6904@mlb.semi.harris.com> dbrillha@dave.mis.semi.harris.com (Dave Brillhart) writes:
- : The PERL-based passwd replacement (described in the PERL book and
- : available on the net) implements password aging by using a few bits
- : in the seed characters that start out the encrypted password.
-
- Lest this worry someone unnecessarily, I should point out that this
- encoding method doesn't negate the effectiveness of the salt characters.
- It actually usurps half of the bits to hold the number of weeks since
- 1970, modulo 64, but it purturbs this number using the first two bytes
- of the login id, so even if everyone on the machine changes their
- password in the same week, everyone still gets a different salt (unless
- you start everyone's login id with the same two characters, of course).
-
- : We use the utility and it works great. Besides password aging, you
- : get a terrific pro-active password checker - it will not allow
- : "easy" passwords (defined by you using REs and dictionaries). It
- : keeps you from having to run crack as often :-)
-
- Thanks for the plug. I agree that pro-active is much better than
- post-active (con-active? amateur-active?). Programs like crack
- don't buy you much unless you enjoy tormenting your CPU.
-
- (Potential pro-active passwd program programmer pedagogoical pointer:
- Don't write a program that dictates what kind of passwords are GOOD--in
- general this greatly reduces the search space for a cracker who knows
- how to generate "good" passwords. Your program should only worry about
- what kinds of passwords are BAD. If you let your users figure out a
- good way to generate passwords, they won't all choose the same way.)
-
- Larry Wall
- lwall@netlabs.com
-