* Mikael Abrahamsson
On Mon, 8 Apr 2013, Tore Anderson wrote:
AIUI, the standards-track flavour of MAP, MAP-E, is *not* NAT - it is tunneling, pure encap/decap plus a clever way to calculate the outer IPv6 src/dst addresses from the inner IPv4 addresses and ports. The inner IPv4 packets are not modified by the centralised MAP tunneling routers, so there is no "Network Address Translation" being performed.
This is all splitting hairs. Yes, the outside port addresses do not change but however the src/dst addresses change (=translated), right?
There is no outside port addresses. The Next Header field in the outside IPv6 header is set to 4 (i.e., what follows next is an IPv4 header). This inner IPv4 header (and the payload following it) is the original one and completely unmodified and not translated/rewritten in any way by the ISP's MAP gateway. AIUI, anyway. So unless you mean that the src/dst address "change" or "are translated" due to the addresses in the outer IPv6 header are not the same as in the inner IPv4 header, there is simply is no translation happening here. If this is to be called "translation", then any tunneling mechanism that works by stacking layer-3 headers, including GRE, IPIP, ESP, and proto-41, must be also called "translation".
Does anyone see MAP-E being implemented on regular linecards or is it going to be implemented on processor based dedicated hardware? At least initially, I would just assume it's going to be some kind of CGN blade.
No idea, sorry. Tore