[11206] in Kerberos-V5-bugs
[krbdev.mit.edu #6574] prompter users should zero out reply data
daemon@ATHENA.MIT.EDU (william.fiveash@sun.com via RT)
Thu Oct 15 14:08:06 2009
Mail-Followup-To: rt@krbdev.mit.edu
mail-copies-to: never
From: "william.fiveash@sun.com via RT" <rt-comment@krbdev.MIT.EDU>
In-Reply-To: <rt-6574@krbdev.mit.edu>
Message-ID: <rt-6574-31822.12.1735092363623@krbdev.mit.edu>
To: "'AdminCc of krbdev.mit.edu Ticket #6574'":;"'AdminCc of krbdev.mit.edu Ticket #6574'":;@MIT.EDU
Date: Thu, 15 Oct 2009 18:04:41 +0000 (UTC)
Reply-To: rt-comment@krbdev.MIT.EDU
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Errors-To: krb5-bugs-bounces@mit.edu
I've been looking at the code that calls krb5_prompter_posix() and I've
noticed that the reply data (which may contain a password) is not being
zeroed out before being freed. When I look at similar code in the
Solaris libpam I notice that the reply data is being zeroed prior to
free. This seems to be a better practice security wise.
I suggest looking at every place a prompter is called and memset() the
reply data with 0 prior to it's being freed.
_______________________________________________
krb5-bugs mailing list
krb5-bugs@mit.edu
https://mailman.mit.edu/mailman/listinfo/krb5-bugs
