RE: Question on EAP statemachine
From: Pasi.Eronen (Pasi.Eronennokia.com)
Date: Wed, 29 Jun 2005 12:38:02 -0400 (EDT)
>From the peer state machine point of view, the current EAP
conversation ends and a new one can begin (on this peer state 
machine instance) when the lower layer clears portEnabled
or sets eapRestart. It's up to the lower layer to define 
when this happens.  

For instance, PANA has a special PANA-FirstAuth-End-Request 
message and a bit in the message header that tells when the 
first conversation is over and the second one begins.

Best regards,
Pasi

> -----Original Message-----
> From: ext Mahesh Kelkar [mailto:mkelkar [at] rocketmail.com]
> Sent: Wednesday, June 29, 2005 7:15 PM
> To: Eronen Pasi (Nokia-NRC/Helsinki); eap [at] frascone.com
> Subject: RE: Question on EAP statemachine
> 
> 
> Pasi,
> 
> Great!! thanks. How do you define a single EAP
> conversation? I mean when would you say that EAP
> converation is started? and when would you say that EAP
> conversation is ended?
> 
> Thanks
> Mahesh
> 
> --- Pasi.Eronen [at] nokia.com wrote:
> 
> > Mahesh,
> > 
> > Using multiple authentication methods in a single
> > EAP conversation is explicitly prohibited by RFC 3748.
> > 
> > So if multiple authentication methods are used, they 
> > have to be in separate EAP conversations (like in PANA),
> > and the lower layer has to be aware of these. (Unless
> > a "tunneled" method is used; but that's a single 
> > authentication method from RFC 3748 point of view.)
> > 
> > Best regards,
> > Pasi

Results generated by Tiger Technologies using MHonArc.