On 23.11.2014, at 23.14, Michael Richardson <[email protected]> wrote: > I have read: > Automated Delegation of IP6.ARPA reverse zones with Prefix Delegation > draft-andrews-dnsop-pd-reverse-02 > > as a method to delegate reverse zones to CPE devices as the prefix > is delegated. I find the method entirely sensible, and I think highly > secure. I don't know if this belongs in dnsop or in homenet (or dhcpv6, > since a new DHCPv6 option is requested, and this enhances DHCPv6-PD): I'll > let the INT and Ops ADs sort that out.
Is this actually desired by the operators? At least here (.fi), ISPs seem to consider the reverse pointing to x.customer.y.isp.fi a feature, not a bug, of the current IPv4 deployment and specified same for future IPv6 deployments as well. (At the moment my ISP does not officially support IPv6, but do provide it via 6rd, and I get NXDOMAIN for reverses, but v4 is populated for more or less all ISPs.) > I suggest that one of these WGs should adopt it, and even suggest that this > document should Updates: 6204/7084. If I had the required code point, I > would implement it today in an IPv6 ACS I work on (ServPOET), and contribute > code to Barrier Breaker OpenWRT (to dnsmasq) to do the client end. odhcp6c is the DHCPv6 client there, so some (slightly awkward) integration between dnsmasq/odhcp6c would be implied.. (dnsmasq is relatively statically configured no-IPC daemon, so I suppose odhcp6c could provide it with some sort of config file fragment which provides the information from step 2 in the draft to dnsmasq..) Cheers, -Markus _______________________________________________ DNSOP mailing list [email protected] https://www.ietf.org/mailman/listinfo/dnsop
