Minutes IETF104: intarea
minutes-104-intarea-00

IntArea WG Agenda
IETF 104  - Prague
13:00-14:30       Tuesday Afternoon session I, Room Zurich D

Chairs:
Juan Carlos Zuniga (SIGFOX) - (JCZ)
Wassim Haddad (Ericsson) - (WH)

Minutes taken by Ron Bonica:
======================

1) Discovering Provisioning Domain Names ad Data - Tommy Pauly

 - New RA Option for Provisioning Domains
 - Request for WG Last Call

 Erik Kline - Reads comments that he has posted in email; author agrees
 Mikael Abrahamsson  - Thinks the work is important but wonders if we
 understand implications. Have we experimented with captive portals? Tommy
 Pauly - yes, we have done this in a previous hackathon Erik Kline commits to
 review More than a few people agree to review

  Chairs and Suresh: Wait for secdir review before WGLC

2) SOCK Protocol V6 - Vladimir Olteanu
-   No requests for action

3) Probing IP Interfaces by Virtual Function Index - Ron Bonica

   Extension to probe RFC 8335 (useful when probed interface is unnumbered or
   unreachable) Adding another entry to the registry for VFI Asking for an
   adoption call, very short and simple draft Eric Vyncke: nice to see RFC 8335
   implementation as it is useful for LLA à la RFC 7404

   Michael: How do you determine if the interface is up?
   Ron: Just IF status. There's no guarantee if the ping comes in the interface
   already, it's just the status.

Reviewers: Eric Vyncke and Fred Templin

4)  Generic multi access GMA Convergence Access Protocols - Jing Zhu

    - Suresh (from Jabber) 3GGP document is not a TS (Stage 3)
    - Suresh (from Jabber) It is not accurate to say that 3GGP work is blocked
    from IETF
     - Jing - IETF will need to work on this. 3GGP doesn't do IP protocols.

     Author Request:
        - interested parties to collaborate
       - WG Adoption

    Tom Herbert - Have you looked at GUE?
    Jing - Yes, we want to avoid layer 4 tunneling
    Tom - we added UPD to avoid ACLs. I'm not sure you want another transport
    protocol. Tom - Why did you go with a trailer based protocol? Jing - we did
    this to avoid adding another IP header David Black - Let's have a
    discussion about data paths. Receiver calculates MAC and compares with
    trailer. ESP is bad analogy.
         It's easier if the sequence number comes first.

     Suresh (from Jabber) - Given that MMAS is not on the IETF stream.....

5) GUE Update - Tom Herbert
    - David Black - need discussion of draft-ietf-intarea-tunnels
   - David Black - OK with solution to UDP checksm problem

    - David Black - Look at the Geneve drafts for checksum considerations
   - Gorry Fairhurst - From a congestion considerations perspective, think
   about whether you will deploy in controlled environments or in the wild -
   David Black - 8086 gives you a framework for checksum /IPv6 considerations -
   Mirja - Wy do we need private data?

   Suresh - Insufficient justification for extensions. Can you trim?
       Why should this not be experimental?
   Gory Fairhurst - The GUE documents can't be understood without reading them
   all
               What problem are we solving? Do we really need all of the
               options?
    Fred Templin - We are using GUE in four environments and its being used
    Suresh - That doesn't justify all of the extensibility in the protocol
    Tommy Pauly - Core document should stand alone. Number of degrees of
    extensibility is concerning. Mirja - Agrees with Suresh and Tommy. It's not
    clear why you have so many ways to extend. Fred - As document shepherd, do
    we have a way forward Fred - Can I shepherd extensions, too
   Suresh will send in AD evaluation

6) Guidelines for Registration Procedures for Interface Types - Dave Thaler

Mikael Abrahamsson - Any reason we shouldn't take the expert advice? Let's
trust Dave David Black - Should there be some effort to get all of the tunnels
that we have defined into the registry Suresh - Will AD sponsor draft and start
tunnel roundup

==================================================================