Le jeudi 19 avril 2007 16:25, Gunter Van de Velde (gvandeve) a écrit : > I am slightly confused here. The NAP document is a document to > describe that what you have with NAT can be done natively with > IPv6 (without the header modification option) and this is seen > in both the abstract and the introduction. Problem is: many people understand this as a statement from IETF that CPE should have stateful filtering, and have it enabled by default. So now, people are implementing this in their IPv6-capable CPEs. And then, we realize this re-introduces the connectivity limitations of IPv4+NAT into IPv6. Arguably, this destroys much of the advantage of IPv6 over IPv4 in the unmanaged network scenario. So... yes we certainly need another document to clarify things, and to provide an answer to connectivity issues within IPv6 context (whether that's no firewall, firewall signaling, hole punching, or whatever). -- Rémi Denis-Courmont http://www.remlab.net/
Attachment:
pgpWR9ntnA5o3.pgp
Description: PGP signature