Capwap Mailing List: Re: WTP IPvX address

[Sudhanshu (sudhanshu.ietf]

We should make the protocol behavior identical. It will help in

* Address V6 WTP and V4 AC scenario
* Avoid sending V4 IP address by v6 WTP - a kind of hack

_Suds

-----Original Message-----
From: Pat Calhoun (pacalhou) [mailto:pcalhoun [at] cisco.com] 
Sent: Monday, April 16, 2007 3:21 AM
To: Sudhanshu; capwap
Subject: RE: [Capwap] WTP IPvX address

I believe the idea is that there will not be any IPv6 NATs out there (or
at least there aren't any v6-v6 NATs). The WTP IPv4 Address message
element MUST be included by the WTP to allow the AC to determine whether
a NAT exists between the two nodes. Having it as optional makes it
impossible for the AC to make this determination, and the WTP would
obviously never know if a NAT is used.

The question is whether the WTP IPv6 Address message element would be
useful in a v4-v6 NAT scenario. I suppose it would be useful if the WTP
was on a v6 network, and the AC was on a v4 network. In such a scenario,
the AC would expect to see the WTP IPv4 Address message element, which
would obviously be absent since the WTP is an IPv6 node. So either the
AC rejects the request, or we add the proposed WTP IPv6 Address message
element to make the protocol handling identical, regardless of the
underlying IP version number.

Thoughts?

PatC 

-----Original Message-----
From: Sudhanshu [mailto:sudhanshu.ietf [at] gmail.com] 
Sent: Thursday, April 12, 2007 10:17 AM
To: 'capwap'
Subject: [Capwap] WTP IPvX address

Currently we have not defined the WTP IPv6 address message element. 

WTP IPv6 address message element is not needed as it is used exclusively
for NAT? Or do we assume that WTP will not be IPv6 host?

If former, "WTP IPv4 address message element" should be optional in join
request. If later, what is rational behind this assumption?


-Suds

_________________________________________________________________
To unsubscribe or modify your subscription options, please visit:
http://lists.frascone.com/mailman/listinfo/capwap

Archives: http://lists.frascone.com/pipermail/capwap