Hi Daniel, There are two parts of considerations in dslite, the tunnel part and the NAPT part. Section 7.2 is on the tunnel consideration. The network must not interference the VPN packets. Section 8.2 is on the NAPT in the AFTR. AFTR must support all the requirements in rfc4787 and rfc5382.
We are talking about the host behind B4 element trying to establish a VPN tunnel. Regards, Yiu On 5/3/11 6:35 PM, "Daniel Roesen" <[email protected]> wrote: >Hi Yiu, > >On Tue, May 03, 2011 at 10:15:31PM +0000, Lee, Yiu wrote: >> This change is to address a comment from IESG. Our goal here is to >>ensure >> dslite which is a NAPT44 solution should work like what NAPT44 is >>supposed >> to do today. The word "interfere" should be interpreted that dslite must >> not change any existing VPN solutions for NAPT44. It doesn't mean dslite >> MUST NOT modify the packet. For example: what defined in RFC3947 and >> RFC3948 must not be interfered by dslite. > >Thanks. Shouldn't 7.2 then go under chapter 8 "NAT considerations"? >Aren't VPNs traversing the DS-Lite NAPT44 function covered via 8.2 >anyway, implicitly? > >DS-Lite specific properties and working except the NAPT44 function are >clearly defined, so where's the room for implementors to infer with >VPNs? Unless I'm missing something, 7.2 is quite redundant to 8.2 > >BTW, are we talking about CPE-terminated IPv4 VPN tunnels, or IPv4 VPN >tunnels traversing a B4 CPE (read: host behind B4 CPE trying to >establish a VPN tunnel to some VPN hub "out there")? > >Best regards, >Daniel > >-- >CLUE-RIPE -- Jabber: [email protected] -- dr@IRCnet -- PGP: 0xA85C8AA0 >_______________________________________________ >Softwires mailing list >[email protected] >https://www.ietf.org/mailman/listinfo/softwires _______________________________________________ Softwires mailing list [email protected] https://www.ietf.org/mailman/listinfo/softwires
