Minutes for DNSOP at IETF-90
minutes-90-dnsop-4

WG:                        DNS Operations (dnsop)
Meeting:                IETF 90, Toronto
Location:                Fairmont Royal York, Ballroom C
Date:                        Tuesday, 22 July 2014
Time:                        0900-1130 EDT/1300-1530 UTC
Chairs:                Tim Wicinski <tjw.ietf@gmail.com>
                        Suzanne Woolf <suzworldwide@gmail.com>


0) Introduction  


1) Agenda Bashing, Blue Sheets, etc. (10 min)


        Housekeeping, Charter
        How's Our Driving?


2) Status Updates (15 min)


        - Special Names, Woolf 
Andrew Sullivan: Liason statement shoukd help drive direction of the
special names discussion


        - DNS Privacy, Wicinski


        - draft-ietf-dnsop-dnssec-key-timing, Mekking (5 min
                Outcome: Do we need a formal WGLC?
Stephane Bortzmeyer: found some editorial issues but minor. question on
Tgen step,
implies both keys are generated at same time.
Matthijs Mekking: It appears they are generated at same time, but not
true. Tgen state is useful
for implementors. Would like to keep Tgen, maybe change the text. 

        - draft-ietf-dnsop-resolver-priming, Koch (5 min)
                Outcome: Need a Shepherd to work with chairs
Needs Reviewers: joe abley, warren kumari sign up
Shepherd request: Joe Abley

        - draft-ietf-dnsop-edns-tcp-keepalive, Wouters
- draft-ietf-dnsop-edns-chain-query, Wouters
                 Outcome:  Are we ready for WGLC?
The Chairs forgot to bring these up. 


3) Old Business (20 min)


        - draft-ietf-dnsop-dnssec-roadblock-avoidance, Gudmundsson (10)
                Outcome: Ready for WGLC?
Olafur presenting.
Comments: 
Dan York: says the bad hotel proxy problem means may not know one
resolver is bad
should we document and monitor behavior from proxies? 
Ed Lewis: proxies exist to handle edge cases
Olafur: also bad hotel, enterprise
Phil Halm-Baker: good cataloging everything
Dan: Are you waiting to rev document based on guidance
Olafur: If the guidance is such
Action: document these issues
<unknown>: could as112 be used for test infrastructure
action: move forward on mailing list

        - draft-mglt-dnsop-dnssec-validator-requirements, Migault (10)
                Outcome: Ready for adoption?
Paul Hoffman: Not nearly enough discussion on mailing list. 
*action; to take to list to adopt*
andrew sullivan: do the work people
Volunteer reviewers: Joe and Andrew

4) New Business


  Topic Area:  Optimizing Distribution of root zone data 
        - draft-wkumari-dnsop-dist-root, Hoffman (15 min)
Paul Hoffman: -02 draft coming out. 
Andrew Sullivan: method to filling caches, anbd that exists as TTLs
Lars: Look at Refresh time
Paul: Happy to be more conservative. We wont make certain statements
because
we don't want to do analysis
Joe Abley: Solution looking for problem
George Michaelson: Quality on saying 'how'. Document 'how' and
technologies
Grow some skin! and have a good ol' fight
Peter Koch: goal is limit traffic, 
Paul Hoffman: reducing NXDOMAINs at root


        - draft-lee-dnsop-scalingroot, Lee, Vixie (15 min)
Warren: only affect customers who use you as resolver, vs affecting
other customers.
(very quiet). also avoids conflict.
Jon Dickinson: anycast with 100s of root servers
David Conrad: Reminds him of Ohta's draft from 2004. 

        - dns-authority-server-placement, Ning (10 min)
Mark: one detail: look at topology instead of geography,and an area of
work
John Levine: Lots of work in NP-complete programmiung
Paul Hoffman: draft not about root.3 documents on 3 topics. math in this
draft could affect 
other drafts. Focused on new TLD operators,
Should we "scale stuff"
Andrew Sullivan: Placing of intrastructure
Peter Koch: Like to see more of this. OARC is a better venue perhaps?
Russ Mundy: to be most useful, address geography and topology

Outcome: Where does this discussion take us?

Peter Koch: Focus on problem statement
Suzanne: There may not be a single problem statement
Geoff: No Single Problem Statement.
Wes Hardaker: Problem Statement is "Scale Stuff"
<unkn>: There is not a 'problem statement', but there is a "opportunity
statement"
Dan York/Mark Andrew: Add a Zone Signature. 

Paul Hoffman: concern that WG has consensus on these documents could be
inferred as DNSOP telling root 
zone people what to do.  

 Joel Jaggeli: Will be a decision if appropriate to the charter. Has a
problem with "this is a problem and we think someone else should work on
it"
 
 Humming is guidance for the chairs

Topic: 
        - draft-mekking-dnsop-kasp, Mekking (15 min)
Question: topic to tackle?
<unk>: What are you looking to standardize
Matthijs: Refresh parameters
       
Outcome: An effort to help standarize policy model


        - draft-fujiwara-dnsop-poisoning-measures, Fujiwara (10 min)
                Outcome: 


        - draft-howard-dnsop-ip6rdns, Howard (10 min)


        - draft-jabley-multicast-ptr, Abley (5 min)
Outcome: With limited v6 deployment, do we have a chance 
to not screw this up? 


Other Business: