Consider the case of a mobile user.  It is not sufficient for the user to 
identify himself using his current IP address.  Therefore some other 
identification means is necessary, perhaps in the form of NAI.  Whatever 
this identification mechanism is will be used in the authentication process 
with the central repository.  But in stating the problem, I want to avoid 
narrowing it down to be the NAI, or some other form, as that should be 
defined when we come to proposing a solution for the problem.

Mike
----- Original Message ----- 
From: Melinda Shore
To: Michael Ko
Cc: [email protected]
Sent: Monday, November 28, 2011 7:29 PM
Subject: Re: [IPsec] Discovery (Was: Preparing a charter change for P2P VPN)


On 11/28/11 5:21 PM, Michael Ko wrote:
> [mk] A "user" is a network node that wants to connect with a peer node,
> preferably in a direct end-to-end connection. (If you can suggest a
> better term than "user" that will cause less confusion, I will use it
> instead.) A "user" may not be "authorized" to connect with all peer
> nodes in the domain.

I still find this very unclear.  If a user "has" an IP address, how
is the user identified, and how does the network know who it is?  That
is to say, there's apparently some sort of identity process/
authentication going on here prior to an IKE exchange, and I
cannot tell from your requirements (and frankly I find them too
vague to be called requirements, really) what that identity is,
who's participating, how it's transacted, or what sort of token/
credential/whatever represents that identity, let alone how it's
actually going to be used as the basis for authorizations.  Then,
there's the whole question of how you're going to authorize this
stuff.

I don't think you've got anything in here that's specific enough
to be charterable.

Melinda
_______________________________________________
IPsec mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ipsec

Reply via email to