[tac_plus] Re: Unified passwords?

[john heasley]

Thu, Feb 26, 2009 at 01:54:57PM -0700, Schmidt, Daniel:
> I believe you require the enable patch.  I'm not exactly sure why this
> was not integrated into the last release.  My C skills are not nearly as
> good as John's as to allow me to say that it should go in.  However, I
> use it and it works fine.  Look at the code, it's a very small change.  

I can't just commit something without investigating what side effects it
might have, updating comments, and updating manual pages.  These things
take time that I dont always have.

anyway, this seems to be fine and I've committed it, except for the log
msg changes which I do not understand.

> diff -ruN tacacs+-F4.0.4.15/pwlib.c tacacs+-F4.0.4.15.new/pwlib.c
> --- tacacs+-F4.0.4.15/pwlib.c	2007-12-13 20:18:39.000000000 +0100
> +++ tacacs+-F4.0.4.15.new/pwlib.c	2008-04-22 11:05:11.000000000
> +0200
> @@ -206,7 +206,7 @@
>      /* Oops. No idea what kind of password this is. This should never
>       * happen as the parser should never create such passwords.
>       */
> -    report(LOG_ERR, "%s: Error cannot identify password type %s for
> %s",
> +    report(LOG_ERR, "%s: Error cannot identify password type '%s' for
> '%s'",
>  	   session.peer,
>  	   cfg_passwd && cfg_passwd[0] ? cfg_passwd : "<NULL>",
>  	   name ? name : "<unknown>");
> @@ -260,10 +260,22 @@
>  	return(data->status == TAC_PLUS_AUTHEN_STATUS_PASS);
>      }
>  
> +    p = tac_find_substring("file ", cfg_passwd);
> +    if (p) {
> +        if (!passwd_file_verify(name, passwd, data, p)) {
> +            data->status = TAC_PLUS_AUTHEN_STATUS_FAIL;
> +            return(0);
> +        } else {
> +            data->status = TAC_PLUS_AUTHEN_STATUS_PASS;
> +        }
> +      
> +        return(data->status == TAC_PLUS_AUTHEN_STATUS_PASS);
> +    }
> +
>      /* Oops. No idea what kind of password this is. This should never
>       * happen as the parser should never create such passwords. */
>  
> -    report(LOG_ERR, "%s: Error cannot identify password type %s for
> %s",
> +    report(LOG_ERR, "%s: Error cannot identify password type '%s' for
> '%s'",
>  	   session.peer,
>  	   cfg_passwd && cfg_passwd[0] ? cfg_passwd : "<NULL>",
>  	   name ? name : "<unknown>");
> 
> -----Original Message-----
> From: tac_plus-bounces at shrubbery.net
> [mailto:tac_plus-bounces at shrubbery.net] On Behalf Of Jason Frisvold
> Sent: Thursday, February 26, 2009 12:25 PM
> To: john heasley
> Cc: tac_plus at shrubbery.net
> Subject: [tac_plus] Re: Unified passwords?
> 
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> john heasley wrote:
> >                   user = bart {
> >                       enable = <password_spec>
> >                   }
> > 
> >               enable specifies the enable password.  The
> <password_spec>  may
> >               only  be  of  type cleartext or des.  If the daemon was
> compiled
> >               with per-user enable support (--enable-uenable), the
> host enable
> >               password will be evaluated iff the user does not have a
> personal
> >               enable password.
> 
> I was looking to use /etc/password for both login and enable passwords.
>  If I understand your reply, you're saying that's not possible, correct?
> 
> - --
> - ---------------------------
> Jason Frisvold
> Network Engineer
> frisvolj at lafayette.edu
> - ---------------------------
> "What I cannot create, I do not understand"
>    - Richard Feynman
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.5 (GNU/Linux)
> Comment: Using GnuPG with CentOS - http://enigmail.mozdev.org
> 
> iD8DBQFJpuykO80o6DJ8UvkRAjYJAJ0SDzbt80Y5rGuIVAFNJXSjlIsY6gCfZv7r
> zUWW6YDfvhIYdvE164K3O8g=
> =NAId
> -----END PGP SIGNATURE-----
> _______________________________________________
> tac_plus mailing list
> tac_plus at shrubbery.net
> http://www.shrubbery.net/mailman/listinfo.cgi/tac_plus