Hi Jari,

I agree with you that a more optimized implementation can consume less
amount of memory allocation even when a cookie and the first EAP
request are carried in the handshake phase messages.  In fact, the
specification does not prohibit the operation:

"
   The initial EAP Request message MAY be optionally carried by the
   PANA-Start-Request (as opposed to by a later PANA-Auth-Request)
   message in order to reduce the number of round-trips.  This
   optimization SHOULD NOT be used if the PAA is desired to be stateless
   in the handshake phase since transmission of an EAP Request message
   creates a state at EAP layer.  See [RFC4137] for more information on
   the EAP state machine and the allocation of state information in the
   respective protocol steps.
"

Or are you suggesting to *always* carry EAP message in the handshake
phase messages regardless of L-flag?

Regards,
Yoshihiro Ohba



On Tue, Oct 03, 2006 at 05:26:40PM +0300, Jari Arkko wrote:
> Yoshihiro,
> > Since PaC and PAA may be communicating over multiple IP hops (and thus
> > PAA is exposed to attackers in the Internet in general), it is
> > important from security perspective to support stateless operation
> > until rechability of the PaC from the PAA is verified as much as
> > possible.
> >
> > Speaking of EAP layer state, I know of an existing EAP authenticator
> > state machine implementation that requires more than 500 bytes of
> > memory allocation per EAP session.
> I'm not sure this is a major problem. I agree that we need to
> verify that the client is a real entity before engaging in a
> costly exchange. But the first EAP message is EAP Identity
> Request, and it comes from the PAA, not from the AAA.
> There's really very little need for state at that time other
> than the ID value from the EAP packet, but even this could
> be either static or somehow related to the Cookie value. So
> I think implementations can be built to be stateless even
> if you send the first EAP message at an early stage.
> 
> --Jari
> 
> 

_______________________________________________
Pana mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/pana

Reply via email to