... >-----Original Message----- >From: Manfredi, Albert E [mailto:[EMAIL PROTECTED] >Sent: Tuesday, June 12, 2007 16:29 >To: Roger Jorgensen >Cc: IETF IPv6 Mailing List >Subject: RE: Revising Centrally Assigned ULA draft > >> -----Original Message----- >> From: Roger Jorgensen [mailto:[EMAIL PROTECTED] > >> Why are we even talking about ULA-C now? What you want is nicely >> covered by ULA... not ULA-C but regular plain stright forward ULA. > >Largely, but not completely. Think in terms of large platforms that have >several different networks, administered by different entities, but >easily isolated from The Internet or even a larger intranet. ULA-Cs are >convenient to route within the platform, with the knowledge that they >can easily be filtered at edge routers because they are easy to spot.
And filtering on the normal ULA prefix (or ULA-D, if you will) of FD00::/8 is not easy? The advantage in ULA-C is the guarantee of global uniqueness, without "burning" a PI allocation. > >Just like before, with site-local addresses, I'm sure this can be worked >around, no matter what the decision is. Having to go to a RIR and ask >for PIs is a possibility, although what's nice about ULA-Cs is exactly >that you don't have to go to ask anything of anyone! Isn't the whole point of ULA-C that you do, in fact, need to ask someone? /TJ -------------------------------------------------------------------- IETF IPv6 working group mailing list [email protected] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
