Hi Stewart, Your suggestion is absolutely right. Here, I do not mean to propose any solution, but just to share an understanding of ‘backwards compatible’. If some people interest much about this approach, I’d like to join the creation of a draft on this. It should not be so hard to design the core algorithm.
Guangpeng From: Stewart Bryant [mailto:[email protected]] Sent: Tuesday, March 2, 2021 11:51 PM To: Liguangpeng (Roc, Network Technology Laboratory) <[email protected]> Cc: Stewart Bryant <[email protected]>; Toerless Eckert <[email protected]>; Jiayihao <[email protected]>; [email protected]; int-area <[email protected]>; [email protected] Subject: Re: [Int-area] Using ISO8473 as a network layer to carry flexible addresses On 2 Mar 2021, at 14:21, Liguangpeng (Roc, Network Technology Laboratory) <[email protected]<mailto:[email protected]>> wrote: Hi Stewart, Backwards compatible should be always good thing for investment protection. Except “could put your new packet into a IPv6 parser”, another possible approach is the last new forwarder translate new packet to IPv6 packet (may encapsulate new fields in extension header), and passthrough in IPv6 domain. Then the first new forwarder adjacent to IPv6 forwarder must recognize IPv6 encapsulation and traversing all extension headers to form new packets if next hop is new forwarder. If you could write down the two formats and describe the mappings that would make people reading this thread feel a lot more comfortable that this is feasible. It would also enable everyone to understand any limitations and constraints with the approach you propose. - Stewart Guangpeng Li From: Stewart Bryant [mailto:[email protected]] Sent: Tuesday, March 2, 2021 9:53 PM To: Toerless Eckert <[email protected]<mailto:[email protected]>> Cc: Stewart Bryant <[email protected]<mailto:[email protected]>>; Jiayihao <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]>; int-area <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]> Subject: Re: [Int-area] Using ISO8473 as a network layer to carry flexible addresses On 1 Mar 2021, at 15:33, Toerless Eckert <[email protected]<mailto:[email protected]>> wrote: I really don't understand why the IPv6 world has not understood how the most easy way to allow for the applicability of IPv6 to grow (especially beyond "just more addresses thn IPv4") would be to come up with a backward compatible encap on the wire that would support additional address lengths. Toerless I don’t think there is a simple backwards compatible approach, but we can probably do more than we do today. Backwards compatible means that you could put your new packet into a IPv6 parser and it would correctly forward the packet as if nothing had changed. You could I suppose put a well known IPv6 address in the IPv6 header and put the real address in an extension header, perhaps including the pointer to the address in the suffix of the IPv6 address to make finding the EH much faster, but I am not sure that is backwards compatible. I suppose it might be able to do a bit better if the address in the IPv6 DA was the DA of the egress router and old routers did best effort to the egress and newer routers knew to take a look at the extension header for more detail. I think that it is worth thinking about how we could do better than we do today, but I think we need to be careful with the term backwards compatible. - Stewart
_______________________________________________ Int-area mailing list [email protected] https://www.ietf.org/mailman/listinfo/int-area
