On Thu, Jan 31, 2002 at 11:23:34AM -0500, Nicolas Williams wrote:
[snip -- most of your response snipped, because it is clear that
 we simply disagree.  in fact, my response here is laregly tangential.]
> On Thu, Jan 31, 2002 at 09:42:51AM -0600, Jacques A. Vidrine wrote:
> > On Thu, Jan 31, 2002 at 09:56:47AM -0500, Nicolas Williams wrote:
> > = Even with preauthentication, you can still accumulate ciphertext
> >   _without_ eavesdropping unless the administrator has taken special
> >   precautions: precautions which (like preauthentication) are not the
> >   default policy in new installations of MIT, Heimdal, or Windows 2000
> >   KDCs.

By the way, someone dropped me a line to inform me that
preauthentication is on by default with Windows 2000.

> The original poster wanted to
> know whether Kerberos is more secure than NIS. 

I admit that I was not responding to the original poster, but rather to
your comments about dictionary attacks not being practical versus
Kerberos 5.

> > The word `preauthentication' implies that the AS-REQ message
> > authenticates the user to the KDC.  Neither SRP nor PDM provide
> > authentication to the KDC in a two-message exchange.
> 
> Ah, yes. True. Multi-round-trip pre-auth is acceptable to me and nothing
> in RFC1510 prevents multi-round-trip pre-auth designs.

You miss the point again :-) With SRP or PDM, I do not believe
that preauthentication is necessary.  The primary motivation
[1] for preauthentication was to prevent the type of attack in
which the attacker requests TGTs en masse.  Using SRP or PDM as a
preauthentication method prevents this same attack from succeeding
even with just a two-message exchange, because the messages of such
protocols are not susceptible to a dictionary attack. i.e. the
attacker can still request and intercept messages all day long, but
the messages don't contain any known plaintext encrypted with a key
derived from the password and are therefore useless.

Cheers,
-- 
Jacques A. Vidrine <[EMAIL PROTECTED]>                     http://www.nectar.cc/
NTT/Verio SME           .      FreeBSD UNIX      .        Heimdal Kerberos
[EMAIL PROTECTED]      .   [EMAIL PROTECTED]   .           [EMAIL PROTECTED]

[1] There are other useful features of preauthentication that would be
    lost with a two-message SRP/PDM preauthentication method.  The KDC
    cannot then distinguish between successful and failed attempts, so
    implementing `last login' or `X failed login attempts' fields at the
    KDC is no longer possible.

Reply via email to