[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: pam_{unix,pwdb}: crypt/md5 necessary?

On Wed, 2 Aug 2000, Michael Tokarev wrote:

> In the pam_{unix,pwdb} modules, there are an implementation of md5,
> bigcrypt etc.  Just for curiocity: are them needed on linux (or, on
> system with glibc) where "old good" crypt already handles md5 passwords
> via "magic" salt?  If I remove (conditionally exclude) those routines
> from that modules -- will it be ok?  And what condition should be
> tested (i.e. what preprocessor define) for this?
> (I think that implementing this routines on glibc-based system is just
> a waste of resources, not big but anyway...).

It sounds like an OK idea to me.  If you're going to do this, I wouldn't try
to test for preprocessor defines at all.  The only real test, AFAIK, is to
compile a small sample program and run crypt() with an md5 salt, then check
the output.

Steve Langasek
postmodern programmer

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index] []