Coordinates

• H.323: NIFF/HUNGARNET MCU GDS: 0036100309901
• Phone: +36 1 666 1111 (type GDS number 0036100309901) https://wiki.voip.niif.hu/index.php/En-faq-pstndialin

Attending

• Jason Zurawski (Internet2)
• Roman Lapacz (PSNC)
• Sławomir Trzaszczka (PSNC)
• Piotr Pikusa (PSNC)
• Aaron Brown (Internet2)
• Jeff Boote (Internet2)
• Cándido Rodríguez (RedIRIS)
• Freek Dijkstra (SARA)
• Nina Jeliazkova (BREN)
• Michael Bischoff (SurfNET)

Agenda

• Review actions from last meeting
  • COMPLETE - Jason: Migrate documents to Gridforge
  • COMPLETE - Jeff: Make doodle for next meeting
  • NOT COMPLETE - Everyone: Read current documents, offer comments, suggest additions and changes before next meeting.
• Discuss Document Plans/Hand out assignments - JZ
  • NMC-Base_TODO
  • NMC-MA_TODO
• Result Codes (See Slawomir's mail, 1/12/2010 9:58 UTC, and attachment below) - ST/JB
• OGF 28 Plan - JB/MS
• Next Meeting

Notes

NMC-Base Discussion

Jason outlined the work breakdown (see NMC-Base_TODO). Each section was discussed.

Motivation section (based on Freek's mail) may need some work. Roman mentioned there is someone from GN3 (Nordunet?) that is working on a project to compare netconf to perfSONAR. Roman will send details when he finds them. Jason will also add Jeff and Freek as people to help with the motivation section.

Jason will add Aaron to the chaining section. Aaron/Roman (and Jeff) may want to talk based on their implementations of the chaining in the services. Chaining section needs to be split between NM and NMC. Martin wasn't present to add input, Jeff thinks that we should keep the schema specific bits (e.g. what could be done) in the NM and the protocol specific bits (e.g. what is done) in NMC. The idea being that the protocol is a specific thing that must be followed to the letter. Roman mentions filter chaining and if its something that belongs to NM or NMC. Something to think about going forward.

Quick mention of the Result code section since we are still undecided on the way forward. Sławomir/Roman willing to help with the section. If we do go foward with option 3 there will need to be some research into how others (e.g. HTTP) are doing things. This is how Jeff started the idea, but it makes sense to look into this more.

NMC-MA Discussion

Jason goes over plans fast since most time should be spent on the base. GN3 developers need to review what is there now (especially storage message) to check for accuracy.

Result codes section in this document may not be needed. Base doc should define everything, and caveats (if any) would be here.

Result Codes

Discussion can and should continue on the mailing list. Seems to be narrowing to 2 vs 3. Some differences of opinion for client vs server development. Jason brings Freek up to speed on where the discussion started and some of the objections that were voiced in the past (e.g. why not use underlying technologies like SOAP or HTTP for error codes). Jeff reiterates that the plan is not to use them as is, but use them as a model. This makes sense to most, but agreement is probably needed.

OGF 28

Jeff requested 2 sessions for NMC (1 for NM). Freek requested 2 for NML. NSI probably has several reserved.

Jeff/Martin/Jason expecting to go. Freek will be going. PSNC does not know if they will go yet, but they are trying.

Next Meeting

The same time seemed ok for all - next meeting in 2 weeks (1/28/2010).

Actions

• All: Review documents if you haven't done so
• All: Review NMC-Base_TODO and NMC-MA_TODO, see if you have an assignment.
• Roman: Find contact information for GN3 contributor looking at netconf vs perfSONAR protocols
• Jason: Set agenda for next time, update notes.
• Freek: Review and correct RFC 2119 words.

Next Meeting

Jan 28th, 2010 - 16:00 UTC

Attachments:

nmc-eventTypes-20100113.pdf [20100114Notes/nmc-eventTypes-20100113.pdf]