EricLKlein wrote: To keep in context this is what Naiming Shen wrote...
> > As long as we are clear that, the "site-local" does not get special > > treatment in terms of routing and dns, we should care less about if > > "site-local" is deprecated or still lived. It's perfectly fine and > > actually somewhat useful if "site-local" plays the same role as in > > IPv4 addresses defined in rfc1918. > > The complete point behind IPv6 and 128 bit addresses is that we will get back what was once normal: End to End communication. And for that one needs globally unique IP's. If you imply site-locals this complete point is gone and then one can better stay with IPv4 and one big InterNAT. If you really want to have IP space for 'non-internet networks' then one should talk to the RIRs and request address space for this. This address space should then be divided in /48's and everyone who has a requirement for IPv6 address space and has a network which is not connected to the internet can request a /48 from that space. This will at least enforce global uniqueness. One should note that when one does connect to the internet this address space is futile as a single /48 cannot be announced into the DFZ and thus it will never ever be routable. Site-locals imply that networks will choose 'easy to remember' addresses and thus will cause dupes. And they won't be globally unique. > Or to ask it a different way, (and maybe this is the > solution) will all of the 10.x.x.x (and the other IPv4 > private addresses) suddenly become globally broadcast? Please define 'global broadcast' > > This is a bigger problem. What will an IPv6 application or > hardware do with a ::10.x.x.x address? >From a NetBSD's netstat -rn output (trimmed a bit): Internet6: Destination Gateway Flags Refs Use Mtu Interface ::/104 ::1 UGRS 0 0 33220 lo0 => ::/96 ::1 UGRS 0 0 33220 lo0 => ::127.0.0.0/104 ::1 UGRS 0 0 33220 lo0 ::224.0.0.0/100 ::1 UGRS 0 0 33220 lo0 ::255.0.0.0/104 ::1 UGRS 0 0 33220 lo0 ::ffff:0.0.0.0/96 ::1 UGRS 0 0 33220 lo0 2002::/24 ::1 UGRS 0 0 33220 lo0 2002:7f00::/24 ::1 UGRS 0 0 33220 lo0 2002:e000::/20 ::1 UGRS 0 0 33220 lo0 2002:ff00::/24 ::1 UGRS 0 0 33220 lo0 In other words: nullroute those 'special' blocks. Same goes for RFC1918 space btw. Afaik there is no BCP about this though. Greets, Jeroen -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
