In message <54be75d7-e70b-46ab-93c1-042e655bb...@apple.com>, Stuart Cheshire wr ites: > On 29 Jan, 2014, at 07:47, Ralf Weber <d...@fl1ger.de> wrote: > > > Where shall this stop? How about .LOKALESNETZWERK (german for .LAN). > > How many domains do we want to treat special? I know this draft only asks > > for 8, but some of them are on ICANNs application list. > > Currently, with no established procedure for local-use names, the result > is chaos. Since no DNS equivalent to RFC 1918 exists, people use whatever > name they feel like. My hope is that if people are offered a short list > of legitimate pseudo-TLDs for local-use names, the temptation to use some > other TLD not on that list will be less. Today all NAT gateways I know of > default to one of the RFC 1918 address ranges. If RFC 1918 did not exist, > would NAT gateways not exist, or would they just hijack who-knows-what > addresses? I suspect the latter.
There is a slight difference. There is no restriction on getting names other than unwillingness or lack of education about how to get a name. There is a real restriction on getting enough IPv4 addresses to avoid having to use NAT. With IPv6 you don't even need split DNS. They can alway use <name>.10.in-addr.arpa. 10.in-addr.arpa is a perfectly legal suffix for a hostname and 10.in-addr.arpa has a insecure delegation so there are no DNSSEC issues. Additionally the AS112 server will soak any leaked names. It's not as pretty as .home or .lan or .corp or .... but it will work. Alternatively get home.arpa insecurely delegated to a AS112 style server. <hostname>.home.arpa. isn't too bad and matches the hostname structure of 99% of the corporate work, <hostname>.<label>.<tld>. This also has the advantage that it doesn't involve root zone politics. Mark > If we acknowledge and document the reality, the IETF can have a role in > guiding it in a sane direction. If we pretend local-use names don't > exist, then the IETF has less relevance in the real world and the real > world carries on without us. > > > I also don't think there are risks in delegation these other than the > applicants will get lots of traffic. > > No, the risk is that the applicants *won't* get the traffic they want, > because some user's local DNS is answering those queries. > > If we have *some* pseudo-TLDs reserved for local-use names, there's a > stronger argument that local hijacking of other names is illegitimate. > > And yes, if you want .LOKALESNETZWERK, then argue for that. Let's use > this IETF discussion process to get some clarity on which names are > local-use and which ones are not. > > Stuart Cheshire > > _______________________________________________ > DNSOP mailing list > DNSOP@ietf.org > https://www.ietf.org/mailman/listinfo/dnsop -- Mark Andrews, ISC 1 Seymour St., Dundas Valley, NSW 2117, Australia PHONE: +61 2 9871 4742 INTERNET: ma...@isc.org _______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
