Op 11 okt. 2012, om 13:50 heeft Ole Trøan het volgende geschreven: >>> PS: I'd also be quite happy to see BRDP go out as an >>> Experimental status RFC initially, whether through >>> the IETF track, IRTF Routing RG, or the ISE-track. >>> BRDP seems very useful and useful protocols tend >>> to self-deploy into environments where they are useful. >> >> We also have Eric Kline's draft in similar space: >> draft-kline-default-perimeter-00. >> >> Both can hopefully contribute to the solution for this problem space in >> homenet. > > I have read the BRDP framework document, but I failed to understand how BRDP > solves the problem of how a border is discovered. > it explains well how to act when a border has been established, but I don't > understand how BRDP helps getting to that point. pointers?
I merge the old BRDP drafts into one, to be submitted soon. Now I have to tweak a bit for Homenet. I do this in slack time, I hope to do some of it before Atlanta. In a MANET and in general, uplink interfaces (wired, cellular, satcom) are distinguished easily from MANET interfaces. In homenet and ethernet, we could have fixed assigned and color marked interfaces, but this is not our target. So here we need an additional tool. After an interface is discovered as an uplink interface, and it has received a prefix (e.g. DHCP-PD), then BRDP propagates this information downstream in the edge network. This could be used for perimeter discovery also. If there is no border router upstream, ask for a prefix with DHCP-PD. If it has assigned such successfully, start acting as BRDP. Other checks would be useful (draft-kline-default-perimeter). I'll add this in the to be submitted homenet-brdp. Teco > > cheers, > Ole > > _______________________________________________ > homenet mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/homenet _______________________________________________ homenet mailing list [email protected] https://www.ietf.org/mailman/listinfo/homenet
