[552] in bugtraq

home help back first fref pref prev next nref lref last post

Re: Sun Patch Id #102060-01

daemon@ATHENA.MIT.EDU (Jon Peatfield)
Wed Dec 21 16:13:35 1994

To: der Mouse <mouse@Collatz.McRCIM.McGill.EDU>
Cc: bugtraq@fc.net, jp107@amtp.cam.ac.uk
In-Reply-To: Your message of "Tue, 20 Dec 1994 06:43:42 EST."
             <199412201143.GAA19647@Collatz.McRCIM.McGill.EDU> 
Date: Wed, 21 Dec 1994 18:24:16 +0000
From: Jon Peatfield <J.S.Peatfield@amtp.cam.ac.uk>

> Kinda sad, because passwd -F is mildly useful, and it's really really
> easy to make it secure: just permanently throw away all elevated
> privilege as soon as the -F is noticed on the command line.  Then
> proceed to run as normal.

Well it may be useful in some environments (we used to use it to maintain a 
proto-password file of allocated users), but it *never* worked properly if you 
had shadow passwords switched on which was kind of sad.  It always insisted on 
looking in /etc/security/ for the password.adjunct which defeats the point of 
having the -F option.  When we heard about the -F security holes we did the 
binary patch thing to remove the -F option.  These days we live without it.

-- Jon

Jon Peatfield, Computer Officer, the DAMTP, University of Cambridge
Telephone: (+44 223) 3-37852     Mail: J.S.Peatfield@damtp.cam.ac.uk


home help back first fref pref prev next nref lref last post