On Jul 29, 2020, at 02:13 , Saku Ytti <saku@ytti.fi> wrote:
On Wed, 29 Jul 2020 at 10:03, Vincent Bernat <bernat@luffy.cx> wrote:
This is the solution Cumulus is advocating to its users, so I suppose they have some real users behind that. Juniper also supports RFC 5549 but, from the documentation, the forwarding part is done using lightweight tunnels.
I'm not sure if you claim otherwise, but no real 'tunneling' takes place, as far as I know, it's internal implementation detail having IPV6 next-hop for IPV4. I don't think there is any additional headers or any additional lookup or cost. Cisco supports extended nexthop encoding too, so it is fairly well supported by shipping products.
In reality, next hop isn’t really a layer 3 address. The layer 3 address is a stand-in that is resolved to a layer 2 address for forwarding. The layer 3 next-hop address never makes it into the packet. As such, the relationship between the destination address family and the next-hop address family is mostly to avoid breaking the brains of humans. Software to handle mixed-address-families in next hop vs. destination should be a relatively trivial difference from software that requires the address families to match. Owen