Maybe this? https://tools.ietf.org/html/draft-huitema-multi6-hosts-02#section-4.2
All the best, Pascal From: rtgwg <[email protected]> On Behalf Of shyam bandyopadhyay Sent: lundi 11 mars 2019 17:14 To: [email protected] Cc: Alissa Cooper <[email protected]> Subject: Fwd: I wonder what is being going on Dear Area Director, Routing Working Group, IETF I am writing to with the intention to resolve the conflict between draft-shyam-site-multi and RFC 8028, RFC 8043 and draft-ietf-rtgwg-pa-multihoming. I had raised the issue to IESG and IAB earlier. Recently IAB has released a report as a reply to my input. I am quoting IAB's response and attaching the file of IAB response as well. IAB report states: On point 2, the IAB understands from a review of the datatracker that the documents are still under review for publication by the ISE. The IAB does not direct the ISE to publish specific documents; the stream’s editorial independence is its key feature. If the ISE does publish the documents, the IAB believes it will join a longer conversation on the topic of source address based routing, and that this would not normally imply any change of status of RFC 8028 or other documents that form part of that conversation. Other relevant work includes RFC 1970, RFC 2461, RFC 4861, RFC 5533, RFC 7048, and any number of related working group discussions. This is just a statement with over simplification without getting into the details of solution for site multihoming. The term 'source address based routing' is nothing new, but how it is supposed to be used to solve what kind of problem is the matter of concern. Traditionally routing is based on destination address and there are multiple protocols like RIP, BGP, OSPF and each has got its own flavor and use. Documents like RFC 1970, RFC 2461, might have used the term 'source address based routing', but none of them specified how it is to be used to solve the problem of site multihoming. Otherwise, what is the use of writing another document like RFC 8028? Let me describe once again how draft-shyam-site-multi has tackled the issue of routing associated with site multihoming: It is neither completely 'source address based routing' nor completely 'destination address based routing', but it is the combination of both. 1. To communicate from one host to another within the same customer site, it is traditional 'destination address based routing'. 2. To communicate from a host of one customer site to a host of another customer site, it is neither completely 'source address based routing' nor completely 'destination address based routing'. To achieve this with only 'source address based routing' or only with 'destination address based routing', it requires the routing table of the entire world to be brought in, which is a very costly approach. To solve this issue it uses 'source address based routing' from the source host to the customer edge router of that customer site and from CE router of first customer site to the destination host using 'destination address based routing'. In short it is described as 'default routing based on source domain of the source address of the outgoing traffic'. This unique solution was first introduced in section 2.4.1 of draft-shyam-mshn-ipv6-07.txt (from which draft-shyam-site-multi was derived). RFC 8028 just elaborates this solution. RFC 8043 and draft-ietf-rtgwg-pa -multihoming are also based on the same solution. I would request IETF to show me a document where this solution was provided earlier. If any such document exists, I withdraw everything that I claim. Otherwise, I would request IETF to revoke RFC 8028, RFC 8043 and draft-ietf-rtgwg-pa-multihoming. They can be reproduced by making a reference to draft-shyam-site-multi showing what they are trying to achieve on top of whatever has already been stated in draft-shyam-site-multi. This is how contribution of the contributors of RFC 8028, RFC 8043 and draft-ietf-rtgwg-pa-multihoming will be properly evaluated. Thanks and regards, Shyam ---------- Forwarded message --------- From: Alissa Cooper <[email protected]<mailto:[email protected]>> Date: Fri, Mar 8, 2019 at 2:05 AM Subject: Re: I wonder what is being going on To: shyam bandyopadhyay <[email protected]<mailto:[email protected]>> Cc: <[email protected]<mailto:[email protected]>>, Adrian Farrel <[email protected]<mailto:[email protected]>>, <[email protected]<mailto:[email protected]>> Hi Shyam, Please direct your comments to [email protected]<mailto:[email protected]>. Thanks, Alissa > On Feb 25, 2019, at 8:27 PM, shyam bandyopadhyay > <[email protected]<mailto:[email protected]>> wrote: > > Hi Alissa, > > I am writing to you as I am not fully > aware of the process of publication of a RFC. > > There was a last call review date on 02-19-2019 for > draft-ietf-rtgwg-enterprise-pa-multihoming. > > I have said many times earlier that the > basic principle based on which RFC 8028, RFC 8043 > and draft-ietf-rtgwg-enterprise-pa-multihoming > are written, i. e. "default routing based on > source address of outgoing packets" was first > introduced on draft-shyam-site-multi. > > I had written earlier to AD Mr. Martin Vigoureux that > draft-ietf-rtgwg-enterprise-pa-multihoming > should make a reference to draft-shyam-site-multi > with an illustration what it is being trying to > achieve on top of whatever has already been stated on > draft-shyam-site-multi. > > Output of the review from Mr. Nicolai Leymann > shows no major or minor issues found and the > draft can be published with some minor editing. > > I get a feel that Mr. Nicolai Leymann is not > aware of the existence of draft-shyam-site-multi. > > So, I would request the reviewer to go through > draft-shyam-site-multi and suggest necessary changes > required based on my input. > > By the way, IETF should consider publishing > draft-shyam-site-multi prior to publishing > documents that are dependent on draft-shyam-site-multi. > > Thanks and regards, > Shyam
_______________________________________________ rtgwg mailing list [email protected] https://www.ietf.org/mailman/listinfo/rtgwg
