Le 2013-02-08 à 12:08, Roland Bless <roland.bl...@kit.edu> a écrit :
> Hi, > > On 08.02.2013 09:31, Rémi Després wrote: > >> Hosts of an IPv6 customer site can have IPv6 addresses assigned, with >> SLAAC or DHCPv6, BEFORE 4rd is activated. > >> If these addresses become in conflict with the CE 4rd address when >> 4rd is activated, renumbering becomes necessary. The objective (*) >> isn't satisfied. (Reason why the CE itself cannot take another >> address is because this address must be *algorithmically derived* >> from an a IPv4 destination by other CEs.) > > Maybe I don't get it, but it seems that mixed usage of non-4rd hosts > and 4rd hosts in the same subnet is not easily possible, since 4rd > cannot use different IIDs than the algorithmically derived ones. > So conflicts are hopefully detected, but cannot be resolved. There isn't such thing than a 4rd host. There are only 4rd CEs (customer-edge routers) in which IPv4-in-IPv6 tunnels are terminated. > This seems to suggest that you better use a different 4rd subnet-prefix > when configuring 4rd use so that all nodes in this 4rd subnet-prefix > use the same method to derive their IID and you avoid conflicts > with other hosts. Wouldn't it be possible to use those two subnets in > parallel, then there would be no need to renumber existing hosts. Using a 4rd-specific subnet prefix may impose some renumbering when 4rd is activated, which is contrary to our objective. (This prefix, needing to be automatically configured, cannot depend on which subnet prefixes are have been used before 4rd activation.) Regards, RD > > Regards, > Roland > -------------------------------------------------------------------- IETF IPv6 working group mailing list ipv6@ietf.org Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
