Dear WG, With four, admittedly long lasting, drafts now published as RFCs and several others in or approaching WGLC, it is time to look at the remaining action items or milestones for the DNSOP WG. Your chairs would like to devote some time in Quebec to gauge interest in future work. We'd like to identify DNS operations related topics that would call for IETF action _and_ can get enough traction for something to be delivered.
So, here are topics that have been raised in the recent or not so recent past, in no particular order: o AS112 next steps The AS112 operations community, supported by OARC, is thinking about expansion in the direction of IPv6 and automated adjustments to the list of zones served by the AS112 nodes o Trust history See recent discussion on the list. o Child-/Parent- Key (Material) exchange Also recently reappeared in the form of a concrete proposal for a solution. Most of the problem descriptions and/or solutions have in common that the signalling and/or data transport happens in-band as opposed to through a dedicated provisioning protocol. o DNS Operator Change under DNSSEC Apparently identified as a real world problem in the registry-registrar- operator world, part of this has been addressed in RFC4641bis. Part of the solution might be more a provisioning protocol issue than a DNS one. o A/AAAA multiple queries issue Over the years several suggestions have been made to optimize the lookup of A and AAAA RRs for the same QNAME in an attempt to ease or encourage IPv6 deployment. While protocol changes haven't found support (and are out of the scope of DNSOP), it was often enough challenged that there was a problem in the first place. Is there room or need to document current practice and the affect on resolution and server infrastructure? o Updating RFC 1912 RFC 1912 has been a landmark in documenting operational issues and is believed to carry the primary reference to what a "lame delegation" is, amongst other useful things. However, since its publication the DNS operational environment has made significant progress that might o Benchmarking and Performance Measurements Measurements of DNS server software usually report "queries per second" values to judge speed and performance. However, there is no agreed upon set of queries, neither for authoritative nor for recursive servers. o Operational Reality Check for use of DNS in other protocols Several protocols, mostly in the applications area, make, sometimes tacit, assumptions about the DNS, the way "a domain" works or the proximity of "domains" and "networks". Review, Guidance or Reacting to operational challenges might be future routes. o DNS "Signing as a Service" Multiple vendors have started to announce DNSSEC signing services for remote DNSSEC zones, including fetching and pushing zones from and to the customers' DNS infrastructure and also including key management and key material handling (in the direction of the parent). Is there demand for standardization and how much of this is a DNS rather than a provisioning issue? o Name Server Control Protocol Work addressing the requirements as discussed in RFC 6168. Please take this as food for thought, not as an exhaustive list or work proposal. All items would have to pass the admission tests: 0) Is there a problem to solve (or document)? 1) Is there enough momentum to get work done? This includes both editors _and_ reviewers! 2) Is there a reasonable chance to reach consensus while still having a useful result? 3) Is the work item in scope for the IETF and DNSOP? Looking forward to a lively discussion in Quebec! -Peter and Stephen _______________________________________________ DNSOP mailing list [email protected] https://www.ietf.org/mailman/listinfo/dnsop
