On 25 Sep 2013, at 11:17, Sander Steffann <[email protected]> wrote:
> Hi, > >> But I note that the OP is *currently* testing with a single hosts w/ two >> interfaces, so host-based source address selection is what's in play here, >> not (s,d) routing. > > It's not source address selection if the host is replying to an incoming > (ICMP ping) packet. The source address is fixed, and the OP wants the host to > route the outgoing packet through the correct interface. And to complement that the homenet scenario is to ensure the packet gets routed out the right exit to avoid ISP ingress filtering, given a host may not have the information required to perform optimal source address selection for a communication - it's then down to the routing system to push the packet out the right way. We saw at an IETF last year a nice demo of zero configuration OSPF being used to both automatically configure prefixes in a home network, and to route based on source and destination. You can find that code (for Bird) and other interesting stuff here https://github.com/fingon. We also have the zero config OSPF stuff implemented in Quagga, to demonstrate interop (though personally I like Bird). It would be good for people to read and comment on Ole's SADR draft, http://tools.ietf.org/html/draft-troan-homenet-sadr-01. Tim
