2013-02-04 11:37, Fernando Gont <[email protected]> : > On 02/04/2013 06:56 AM, Rémi Després wrote: >> >>> I believe that the original question really was "whether it makes >>> sense to reserve an unused IID range for 4rd". >> >> A 4rd reserved range, for 4rd activation to never require IPv6 >> renumbering, has been for long in the specification studied in >> Softwire. > > But my understanding is that softwires didn't adopt 4rd?
4rd is adopted (on experimental track) under the only condition that 6man confirms that it doesn't conflict with the IPv6 addressing architecture. > If 4rd is published as "experimental", then it doesn't make sense to > permanently assign an address range to it. See (*) below. If 4rd is eventually deprecated, the range will be freed. If it isn't deprecated, the range will remain reserved (which, as you know, is only 1/16384 of available IIDs having u=g=1, i.e. nothing to be alarmed by). If and when IPv4 is deprecated, the reservation will become meaningless. As long as IPv4 isn't deprecated, any solution that has distinctive properties for IPv4 service across IPv6-only clouds, is relevant. >>> As noted by Ran Atkinson, since 4rd is an experimental document, >>> it seems more appropriate to separate a *shared* range for this >>> kind of experiment, rather than assign an IID range to 4rd. >> (*) >> As already answered to Ran: "Making the reservation for one design, >> experimental or not, is a guarantee that no future design with >> conflict with it, experimental or not." If you disagree, please >> explain. > > We have several shared resources (ICMPv6 types, etc.) for experiments. > Why should this be an exception? All other experiment, I suppose, had enough with what has been reserved so far. As far as 4rd is concerned, it needs an IID range that cannot conflict with any host address that complies with RFC 4291. Regards, RD > Thanks, > -- > Fernando Gont > SI6 Networks > e-mail: [email protected] > PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492 > > > > -------------------------------------------------------------------- IETF IPv6 working group mailing list [email protected] Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
