On Friday, September 27, 2013 02:53:06 PM Jim Pingle wrote: > It is only a requirement for SLAAC, yes, but it's also > recommended quite strongly in various RFCs and other > docs from the IETF.
Well, the RFC's haven't always bent themselves toward best practice, just recommendation. And given that we're still all trying to understanding IPv6 in practice, I expect more recommendations to follow in coming years. > The IETF wants /64's everywhere (which IMHO is quite > wasteful, but ...) The argument is that IPv4-thinking in an IPv6 world is counter-intuitive. But the other argument says we always thought 32 bits would be more than enough, just like 640KB of RAM should have been :-). No one really knows how long IPv6 will last, especially if you walk away from today's thinking. > It may work perfectly well for some things, but not > others. I'm not sure I trust everything else to properly > adhere to what _should_ work... :-) 8x years running /112's on LAN's and no issue. There are discussions about ASIC or NPU optimization in route lookups, code that is expecting a certain prefix length, e.t.c. I don't buy it, but that's just me :-). Mark.
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ List mailing list [email protected] http://lists.pfsense.org/mailman/listinfo/list
