On Mar 27, 2014, at 12:39 PM, Doug Barton <dougb@dougbarton.us> wrote:
On 03/27/2014 02:58 AM, Janos Zsako wrote:
Dear all,
people on this list might be interested in one aspect of the proposed policy <http://www.ripe.net/ripe/policies/proposals/2014-01>, <http://www.ripe.net/ripe/mail/archives/address-policy-wg/2014-March/008638.html>,
when it comes to (sub-)allocations smaller than /24. Note that BCP10/RFC 2317 is not necessarily the or the only solution to the problem.
_Technically_ speaking, in my opinion BCP20/RFC2317 is an acceptable solution. It is a trade off: it increases the number of DNS queries,
It doesn't have to increase the number of DNS queries if the parent slaves the child's zone(s). The parent will have the answer for the first query that arrives from a given resolver, and the child will receive subsequent queries directly. Also see https://dougbarton.us/DNS/2317.html for more thoughts. :)
OOC, was any thought given to this approach? http://tools.ietf.org/html/draft-gersch-dnsop-revdns-cidr-01 IIRC, there is a treatise in that draft that relates its proposal to RFC 2317. Eric