I thought that routes in the IPv6 DFZ were not supposed to be more
specific than /32.
Yes, another of the great myths. That somehow all IPv6 addresses would
be PAs, only ISPs would have them, and everyone would route on the
default ISP allocations (/32). Great idea in theory (if all you care
about is keeping route tables small), unworkable in practice if you
look at the demands of the real world (i.e., the last 15 years of
IPv4).
Take a look at draft-narten-radir-problem-statement-00.txt, as it
talks quite a bit about the realities that lead to lots of routing
table entries. The pressures to deaggregate /32s in IPv6 are
fundamentally the same as the pressures that lead to deaggration in
IPv4.
Thomas
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf