[dnsext] Uploaded: edns-client-subnet-00

[Wilmer van der Gaast <[e-mail]>]

Everyone,

A few days ago we uploaded a new version of the edns-client-subnet
draft. It can be found at:
http://tools.ietf.org/html/draft-vandergaast-edns-client-subnet-00

The version number was reset to -00 because of the rename from
edns-client-ip to edns-client-subnet. Without a rename, it would've
been version -02.

Appendix A describes all changes made to the draft since the last
version. In short:

* Rewrote origination sections to clarify that normally only recursive
resolvers generate edns-client-subnet options.
* Discussion on whitelisting or automatically detecting if an
authority supports edns-client-subnet.
* More precisely specifying (optional) transitive behaviour.
* Minor revisions to improve clarity and correctness w.r.t. other RFCs.

Also, since the release of the last draft we have:

* Added support for handling edns-client-subnet options to both our
authoritative nameservers and Google Public DNS, using option code
0x50fa (not officially allocated and *not* to be used permanently).
* Started a small-scale experiment of using edns-client-subnet in real
life, to measure latency improvements and confirm that the idea is
generally feasible. We're working with several third parties (CDNs but
also resolvers) and are looking forward to hearing from anyone else
interested in participating.

We intend to release our findings from the experiment described above
in the next months. In the meantime, comments on the new draft are
welcome as always.


Kind regards,

Wilmer van der Gaast, Carlo Contavalli
Google

Sean Leach
Verisign

Darryl Rodden
Neustar