eap Mailing List: Re: Basic facts about EAP

[Yoshihiro Ohba (yohba]

On Mon, May 02, 2005 at 01:00:05AM -0700, Bernard Aboba wrote:
> > I am not sure if using an EAP method to carry the NAS-ID or port-ID
> > for Channel Bindings is really a good idea as it would actually make
> > the EAP method media-dependent while it would make EAP media
> > independent.
> 
> If the Channel Bindings are treated as opaque blobs, and are represented
> as media-independent attributes, I don't think this creates a media 
> dependency.

When I was talking about channel bindings are treated via EAP methods,
I was referring to draft-arkko-eap-service-identity-auth-01.txt.  It
defines several IEEE 802.11 specific parameters such as BSSID and SSID
with assigning Parameters Identifiers for them.  I see this
media-dependent.  This I-D may be updated to support truely opaque
blobs, but I will still see an issue if all EAP methods that are
capable of deriving an MSK needs to carry the opaque blobs.

Regards,
Yoshihiro Ohba


> 
> For example, NAS-Identifier, Called-Station-Id and Calling-Station-ID are
> all media independent.
> 
> At least the way Channel Bindings are described in RFC 3748, Section 7.15,
> it seems like the idea is to export the Bindings for verification by the
> lower layer.
> 
> This seems like something that needs more explanation, though.  I will
> file an issue on it and try to post some text to the list.