[11789] in Kerberos-V5-bugs
Re: [krbdev.mit.edu #6821] The +preauth default in kdc.conf isn't
daemon@ATHENA.MIT.EDU (D.H.Davis@bath.ac.uk via RT)
Wed Nov 17 11:00:10 2010
Mail-followup-to: rt@krbdev.mit.edu
mail-copies-to: never
From: "D.H.Davis@bath.ac.uk via RT" <rt-comment@krbdev.MIT.EDU>
In-Reply-To: <rt-6821@krbdev.mit.edu>
Message-ID: <rt-6821-33475.15.5025792539068@krbdev.mit.edu>
To: "'AdminCc of krbdev.mit.edu Ticket #6821'":;"'AdminCc of krbdev.mit.edu Ticket #6821'":;@MIT.EDU
Date: Wed, 17 Nov 2010 11:00:08 -0500 (EST)
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
On Wed, 17 Nov 2010, Greg Hudson via RT wrote:
> From: Greg Hudson via RT <rt-comment@krbdev.mit.edu>
> To: D.H.Davis@bath.ac.uk
> Date: Wed, 17 Nov 2010 15:33:13
> Subject: [krbdev.mit.edu #6821] The +preauth default in kdc.conf isn't always
> obeyed.
>
> Prior to 1.8, addprinc -randkey was implemented in three
> RPCs: create the principal with a dummy password and the
> disallow-all-tix flag, randomize its password, unset the
> disallow-all-tix flag. This had the unfortunate side effect of
> ignoring the KDC's default flags.
>
> There is now a better way (create the principal with a null
> password), but clients and servers both have to be at 1.8 for it
> to work.
Thanks for the very prompt reply.
I wondered if something like this was happening when I noticed
-randkey with 1.6.3 and 1.7.1 kadmin produced principals with a Key
vno of 2, whereas -randkey with a 1.8.x kadmin produced principals
with a Key vno of 1.
I note you're using RT so please flag this ticket as closed, if
you haven't done so already. It would be unreasonable to expect
the improved interface to be back-ported to earlier versions of
kerberos.
(This arose because I've recently switched to using the +preauth
default for all principals associated with humanoids. Easily
done by making it the default in kdc.conf. Typically I, and I
suspect others, use -randkey when generating host and service based
principals. So I was expecting to have to turn off preauth on such
principals as I'm not quite ready to go there yet. I was puzzled
when I didn't have to turn off +preauth with a production KDC
running 1.6.3. It had already been done for me!)
--
Dennis Davis, BUCS, University of Bath, Bath, BA2 7AY, UK
D.H.Davis@bath.ac.uk Phone: +44 1225 386101
_______________________________________________
krb5-bugs mailing list
krb5-bugs@mit.edu
https://mailman.mit.edu/mailman/listinfo/krb5-bugs
