| RE: Issue 287 | <– Date –> <– Thread –> |
|
From: Adrangi, Farid (farid.adrangi |
|
| Date: Fri, 4 Mar 2005 05:52:42 -0500 (EST) | |
Hi Geln, all Thanks for your time. Please see my comments inline. -Farid > -----Original Message----- > From: Glen Zorn (gwz) [mailto:gwz [at] cisco.com] > Sent: Thursday, March 03, 2005 7:50 PM > To: Adrangi, Farid > Cc: radiusext [at] ops.ietf.org; gwz [at] cisco.com > Subject: Issue 287 > > > There still seems to me to be a problem here. The thinking > portrayed in the revised document seems awfully fuzzy to me. This > is exemplified by the following: "If the identity hint is not sent > initially (such as when the authenticator does not support this > specification), then if the local AAA proxy/server implementing this > specification receives an EAP-Response/Identity with an unknown > realm, it SHOULD reply with an EAP-Request/Identity containing an > identity hint." Of course, the AAA proxy _cannot_ receive an > EAP-Response, nor send an EAP-Request, since it is a _AAA_ proxy, > not an EAP entity of any variety. Well, it is my understanding that the EAP-aware AAA proxy/server [RFC3579] in the access network can send EAP-Request/Identity within an Access-Challenge. Here is a quote from RFC3579: " Rather than sending an initial EAP-Request packet to the authenticating peer, on detecting the presence of the peer, the NAS MAY send an Access-Request packet to the RADIUS server containing an EAP-Message attribute signifying EAP-Start. The RADIUS server will typically respond with an Access-Challenge containing EAP-Message attribute(s) encapsulating an EAP-Request/Identity (Type 1). " Maybe we can make the paragraph more clear by revising to something like this: " The EAP authenticator MAY send an identity hint to the peer in the initial EAP-Request/Identity. If the identity hint is not sent initially (such as when the authenticator does not support this specification), then if the local EAP-aware AAA proxy/server implementing this specification receives an AAA Request packet with an unknown realm, it SHOULD reply with an EAP-Request/Identity containing an identity hint. For example, in case of RADIUS, if the EAP-aware RADIUS proxy/server [RFC 3579] receives an Access-Request packet with an unknown realm in the UserName(1) attribute, then it can reply with an EAP-Request/Identity containing an identity hint within an Access-Challenge packet. Please see "option 3" in the appendix for the message flow diagram. " What do you think? > This is the same thing that I > wrote about earlier, when I said that this draft is interfering with > the EAP protocol: it seems clear that whatever is sending these > "hints" is an EAP entity of some sort (because it is engaged in at > least a subset of the EAP protocol), but it's not a peer (since it's > not being authenticated), it's not an authenticator (since it's not > doing the authenticating) and it's not a server (since it doesn't > terminate the authentication method. So what is it? > > ~gwz >
- (no other messages in thread)
Results generated by Tiger Technologies using MHonArc.