Page tree
Skip to end of metadata
Go to start of metadata

Weekly  meetings for 2017 Q4

Schedules  for next meetings:

Oct 18 Wednesday :14.00 UTC /7.00 PDT Meeting #50 summary

1.Agenda Bashing

2. MANO and related news

3. eBook on State of MANO in OPNFV - Discussions on content and formats.

summary

Oct  11, Wednesday : 14.00 UTC /7.00 PDT  Meeting #49 and summary

(IRC Only meeting, no GTM this & next week)

For IRC can use can use http://webchat.freenode.net/  

IRC#opnfv-mano

1.Agenda Bashing

2. White Paper discussions - Any proposals or eBook-4-MANO?

3. Follow up from last week.

summary

Oct  4, Wednesday : 14.00 UTC /7.00 PDT  Meeting #48

(IRC Only meeting, no GTM this & next week)

For IRC can use can use http://webchat.freenode.net/   

IRC#opnfv-mano

1.Agenda Bashing

2. September Month end report and Actions from last week

3. Best practices that can be carried forward from E  to F and upstream (ONAP - VNF Model . Usecase def. )

4. White paper for on State of MANO in OPNFV?

summary

Sept 27, Wednesday : 14.00 UTC /7.00 PDT  Meeting #47 and Summary

1.Agenda Bashing

2. VCPE Use case discussions for Opera/ONAP vs. vIMS in OPNFV

3. Action Items to follow from last week.

summary

Sept 20, Wednesday : 14.00 UTC /7.00 PDT  Meeting #46 and Summary

1.Agenda Bashing

a. VNF requirements (Documentation to reference ) 

b. VNF Validation Program - Derived from ICE program

c. VNF SDK - Combined from ICE and OPEN-O for all tools etc.

Review Video - (Minutes 5-15) https://wiki.onap.org/display/DW/Integration+Meeting+Minutes?preview=%2F11930007%2F13598917%2FAutomating+the+onboarding+VNF+and+service+modeling+using+SDC+API.mp4

3. Template based VNF Orchestration  best practices and testing status review   

  • Template based VNF Orchestration  and code review of opera heat ims testingClearwater-vIMS-Testing  as use case- Prakash/Trinath (FUNCTEST/ Rake test and Ruby 1.9.3 issues?) -Continue fixes
  • Opera: Approach to testing via snaps-oo for VNFs  to be reviewed - Opera - Yingjun Li/Prakash - (snap-oo extensions?) -No progress (Abandon?)
  • Any new developments to note from OPNFV or upstream - community - Open

4. VNF Life cycle Management best practices & testing status review

  • Impact of Pharos POD  configurations for VNF scaling  - Prakash/Narinder (discussed for vIMS - should it be moved to XCI debate?) -Open
  • Any new developments to note from OPNFV or upstream -community - Brief review of OpenDev related to VNF at Edge - (Attendees) -Open

summary

Sept 13, Wednesday : 14.00 UTC /7.00 PDT  Meeting #45 summary

1.Agenda Bashing

3. Template based VNF Orchestration  best practices and testing status review    (15 minutes)

4. VNF Life cycle Management best practices & testing status review (15 minutes)

  • Impact of Pharos POD  configurations for VNF scaling  - Prakash/Narinder (discussed for vIMS - should it be moved to XCI debate?) -Open
  • Orchestra:  No Clearwater  Live test in E.0.0 will it move that to F.0.0? - Orchestra -( Unit-test + Mock) - Michael/Giuseppe (concluded not in E) -Close
  • test-wg follow-up: Suggests MANO vIMS tests to be in OPNFV F-release -Prakash/Michael - (Any objections from Orchestra team?) -Close
  • Any new developments to note from OPNFV or upstream -community - Brief review of OpenDev related to VNF at Edge - (Attendees) - Open

Summary

Sept 6, Wednesday : 14.00 UTC /7.00 PDT  Meeting #44

1.Agenda Bashing

3. Template based VNF Orchestration  best practices and testing status review    (15 minutes)

  • Template based VNF Orchestration  and  Clearwater-vIMS-Testing  as use case- Prakash/Trinath
  • Opera: Approach to testing via snaps-oo for VNFs  to be reviewed - Opera - Yingjun Li/Prakash
  • EUAG and XCI:  Auto - workload and Edge Cloud - Tina / Prasad /Trinath
  • Any new devlopments to note from OPNFV or upstream - community

4. VNF Life cycle Management best practices & testing status review (15 minutes)

  • Impact of Pharos POD  configurations for VNF scaling  - Prakash/Narinder
  • Orchestra:  No Clearwater  Live test in E.0.0 will it move that to F.0.0? - Orchestra -( Unit-test + Mock) - Michael/Giuseppe
  • test-wg follow-up: Suggests mano vIMS tests to be in OPNFV F-release -Prakash/Michael
  • Any new developments to note from OPNFV or upstream -community

Summary

Aug 30, Wednesday : 14.00 UTC /7.00 PDT  Meeting #43 Summary

1.Agenda Bashing

2. Actions items to review from Meeting#42 -  Orchestra -( Unit-test + Mock) , Opera , Auto - workload and Edge Cloud

3. Other items that are pending from Meeting#42 - Orchestra - test-wg followup, Opera - None, Auto- EUAG and XCI

4. mano-wg report for August 2017.- Clearwater-vIMS-Testing / Modeling Requirements

Summary

Aug 23, Wednesday : 14.00 UTC /7.00 PDT  Meeting #42 and Summary

1.Agenda Bashing

2. Follow up on ACTIONS from meeting  for Release-E and F

Testing vIMS based on Clearwater (orchestra)

Testing vIMS in Opera

Clearwater IMS VNF test in ONAP 

AUTO progress

3. Discussions on VNF on-boarding using VNFD & VNF Packaging best practices used by mano stacks (refer upstream) -to Continue

    - Need comments from Auto/Opera/Installers and Orchestra projects in OPNFV (Action- TBD)

4. Any updates on Best-practices  for adding Dynamic discovery of Mano components (Bryan/Uli) - ACTION? No inputs to date should we close this? -TBD

Summary

Aug 16, Wednesday : 14.00 UTC /7.00 PDT  Meeting #41 and Summary

1.Agenda Bashing

2. F-Release OPNFV based on ONAP related proposals in OPNFV (Tina)

3. E-Release Progress in OPNFV for Orchestra (D-2-E) (Michael)

4. Discussions on VNF onboarding using VNFD & VNF Packaging   (refer upstream) (discussions)

 5. integration and testing for Juju and orchestra (Clarification on Orchestra & JOID/Juju - Orchestra-15)

Summary

Aug 9, Wednesday : 14.00 UTC /7.00 PDT  Postponed due to Bridge conflict.

Aug 2, Wednesday : 14.00 UTC /7.00 PDT  Meeting #40 and summary

1.Agenda Bashing

2, Release E updates for MS5 and MS6 from mano participating teams ( PTLs)

   a. Orchestra-15 (Michael Paul) ( Two step Builds for MANO - How?)

   b. Functest Integration for VoLTE testing (Debyan/Narinder) / Reviewing MANO Integration tests from OpenBaton for OPNFV.

3. ONAP-Automated OPNFV (updates - Tina/Bryan) - Opera Scope - Helen/Yingjun Li (Invited)

4. Dovetail - Testing gaps in FUT(functions)  for SUT (NFVI+VIM) - Srikanth Vavilapalli - Ericsson (Invited)

5. Any updates on Best-practices  for adding Dynamic discovery of Mano components (Bryan/Uli) 

Summary

opnfv-mano-wg-report-july-2017

July 26th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #39 and summary

1.Agenda Bashing

2. Best Practices - Options to consider MANO orchestration  in E-Release

      a.Two step MANO Orchestration wiki page  (Narinder)

      b. one step work-around for mano-descriptor until  then (Uli)

3. Orchestra OpenIMScore Charm review along with vIMS (Debyna)

4. Release E updates for MS5 from mano participating teams (all PTLs)

   a. Orchestra-15 (Michael Paul)

5. ONAP-Automated OPNFV (updates - Tina/Bryan)

summary

July 19th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #38 and summary

1.Agenda Bashing

2. Test - How to define common test for MANO based on scenario-complex-ncat for general dummy nfvo-vnfm testing in OPNFV

3. Opera Team changes to scope any and need to review? and seek approvals? / ONAP-Automated OPNFV

4. Scenario Descriptor updates on MANO for example1 based on Orchestra..(Narinder/Uli)

5. Best Practices - What to consider in E-Release (for testing based on 2 & 4 above)

Summary

July 12th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #37 and summary

1.Agenda Bashing

2. Orchestra Release plans - JOID/PoDs/Builds/Scenario/Deploy/Test -Orchestra/JOID Teams,

3. Opera revival & ONAP multiVIM requirements  / Laas -Opera Team changes to scope any and need to review? and seek approvals?

4. Scenario Descriptor on MANO use case to choose for it / Sample VNF team inputs.

5. Best Practices - On-boarding , Packaging, LCM (Orchestra vs Opera base on ONAP)

6. Use case discussions from Summit...VoIP Presentation

summary

July 5th, Wednesday : Meeting Canceled due to July4th Holiday week in USA.

June 28th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #36 and summary

1.Agenda Bashing

2. Revise mano Wiki pages for Release D+E(+F) targets + plans, besides  VNF Packaging and On-boarding by participating projects in different releases.(ALL PTLs)

3. Create and Deliver Documents from MANO Projects for providing their approach to MANO - Agree on common items to present or a template addressing EUAG Pain points and VNF Reference presented in Weekly Tech Meeting (.All PTLs)

4. Discuss and provide inputs for .MANO Scenario Descriptor (Ulrich Kleber)

5. Use case discussions for E /F releases for mano Integration.

summary

June 21st, Wednesday : 14.00 UTC /7.00 PDT  Meeting #35 and summary

1.Agenda Bashing

2. Review Wiki updates for Release D+E(+F) completions + plans, besides  VNF Packaging and On-boarding by participating projects in different releases.(ALL PTLs)

3. China Summit summary  for  MANO WG to share -  Multi-Access-Edge, TBD - https://www.opnfv.org/opnfv-summit-2017-event-recap

4. Create and Deliver Documents from MANO Projects for providing their approach to MANO - Agree on common items to present or a template addressing EUAG Pain points and VNF Reference presented in Weekly Tech Meeting (.All PTLs)

5. Discuss and provide inputs for .MANO Scenario Descriptor (Ulrich Kleber)

summary

June 14th, Wednesday : Meeting Canceled due to OPNFV China Summit

June 7th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #34 and summary

1.Agenda Bashing

2. Wiki updates for Release D+E(+F) completions + plans, besides  VNF Packaging and On-boarding by participating projects in different releases.(ALL PTLs)

3. China Summit preparation by  MANO projects and any presentation/ feedback for/from speakers.- MANO WG Project PTLs Invited  to present anything they plan  for Main Summit  (Bryan Sullivan) /  

Beijing Design (Prakash Ramchandran)

4. Create and Deliver Documents from MANO Projects for providing their approach to MANO - Agree on common items to present or a template addressing EUAG Pain points and VNF Reference presented in Weekly Tech Meeting (.All PTLs)

5. Discuss and provide inputs for .MANO Scenario Descriptor (Ulrich Kleber)

summary

May 31st, Wednesday : 14.00 UTC /7.00 PDT  Meeting #33 and summary

1.Agenda Bashing

2. Containers Packaging and On-boarding considerations for nfvo and vnfm + common practices checklist to consider

3. MNAO WG - focus areas to  facilitate Delivering E release priority & plan for F

4. China Summit preparation by  MANO projects and any presentation/ feedback for speakers.

summary

May 24th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #32 and summary

1.Agenda Bashing (Action Item review from last week)

2. Release E  followups on MANO projects plans and inputs from PTLs

  a.ONAP/OPEN-O Harmonization plans /Opera Readiness :  E-Release ( Action Item Closed -See Updates from Opera Yingjun Li and  ONAP acting release manage Gildas on ONAP)

  b.Orchestra has committed for  Build Deploy for E-Release and two JIRA tickets are opened (updates from Giuseppe -Await for his availability)

  1. Orchestra-1
  2. Orchestra-2

  c. Write a Juju charm for installing Open Baton and ONAP / Test Juju charm - (Action Item closed)

 d. Are projects in MANO WG  addressing EUAG Pain Points? Sample VNF is taking on the VNF Null(L2Forwarding), Echo.and what will the VoLTE, vCPE  (Moved back to Tech meetings - Action  Item Closed)

 f. How do we get to work in OPNFV for projects on F-Release those that missed E-Release window mid-cyle ? looking from community inputs and suggestions?

  JOID plans to support OpenStack Ocata with noSDN scenario for OpenBaton in release-E. (Action Item Closed )

  If ODL support for charm is added later it may consider that for OpenBaton at that time and Giuseppe to take it further for OpenBaton/Orchestra))

  d. Review of other projects in MANO WG that have indicated intent to participate in E-Release include Domino, Parser, Models and Copper (refer to f above for closure)

  e. Impact of kubernetes orchestration and containers on MANO Integration (Impact/Status of Helm in E-Release - no plans till Aug 3 and Action Item moved to -  Infrastructure WG -Giuseppe / Uli to start discussion on container scenario with LXD in JOID.

  f. Will any projects use Pike features for OpenStack in MANO integration? -None heard in OPNFV to date.

3. Discussions on Open Source SBC use case from MANO for Design and Deployment - Moved to Models-Domino and irc only meetings on Monday's  14 UTC/6 AM PDT #opnfv-domino

4. Events and planning MANO issues

OPNFV(Beijing) - Action on Demo PoC VoLTE(Moved to JOID meeting)

OPNFV Meetup (WSO2) and APIs (ONOS/ONAP)

5. Continuing discussions on VNF On-boarding, Packaging  best practices for MANO rationalization over OPNFV and inputs on ICE.

summary


May 17th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #31 and summary

1.Agenda Bashing

2. Release E  followups on MANO projects plans and inputs from PTLs

  a.ONAP/OPEN-O Harmonization plans /Opera Readiness :  E-Release (Updates from Yingjun Li on what are the issues for deferrals)

  b.Orchestra has committed for  Build Deploy for E-Release and two JIRA tickets are opened (updates from Giuseppe)

  1. Orchestra-1
  2. Orchestra-2

  c. Write a Juju charm for installing Open Baton / Test Juju charm - (other inputs from Narinder)

  d. Are projects in MANO WG  addressing EUAG Pain Points? Discuss Compile by projects what each project is doing for the same for weekly Tech meetings.

  JOID plans to support OpenStack Ocata with noSDN scenario for OpenBaton in release-E to start with. If ODL support for charm is added later it may consider that for OpenBaton at that time.

  d. Review of other projects in MANO WG that have indicated intent to participate in E-Release include Domino, Parser, Models and Copper

  e. Impact of kubernetes orchestration and containers on MANO Integration (Impact/Status of Helm in E-Release)

  f. Will any projects use Pike features for OpenStack in MANO integration?

3. Discussions on Open Source SBC use case from MANO for Design and Deployment 

4. Events and planning MANO issues

Openstack (Boston) - Reports & OPNFV(Beijing) -plans and OPNFV Meetup (WSO2) and APIs (ONOS/ONAP)

5. Continuing discussions on VNF On-boarding, Packaging  best practices for MANO rationalization over OPNFV

summary

May 10th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #30 and summary

1.Agenda Bashing

2. Events and planning MANO issues

Openstack (Boston) - Reports & OPNFV(Beijing) -plans and OPNFV Meetup (WSO2) and APIs (ONOS/ONAP)

.summary

May 3rd, Wednesday : 14.00 UTC /7.00 PDT  Meeting #29 and summary

1.Agenda Bashing

2. Events and planning MANO issues

Openstack (Boston) & OPNFV(Beijing) and OPNFV Meetup (WSO2) and APIs (ONOS/ONAP)

Due to change in Goto meeting  (https://global.gotomeeting.com/join/229632485 ) the meeting had a break and only two topics were covered.

We will use this new Bridge from next week and is also listed here

Summary

April 26th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #28 and summary

1.Agenda Bashing

2. Release E  followups on MANO projects / PTLs

3. Updates for new LF API Tool - Chintana (WSO2)

4. Prepare inputs  for Dovetail test qualifications for selected use case  from Models. (Model team/Prakash/)

5. Should MANO WG wiki be updated for ETSI updates -Release 2 presented by (Pierre Lynch)

6. Events and planning MANO issues for Openstack (Boston) & OPNFV(Beijing)

Summary

April 19th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #27 and summary

1.Agenda Bashing

2. Release E  followups on MANO projects / PTLs

  • Project PTLs inputs to E-Release content (VNF Onboarding & Deployment focus as Bryan Suggested)
  • Code Review  - Opera/Canonical  team members
  •  TOSCA for MANO scenario descriptor file from last week discussions - Bryan/ Narinder/ Uli - Await inputs
  •  SBC Opensource license inputs from LF
  •  SBC with 3-node VES / Hello-Tacker and VDU BP layout

3. Updates for new LF API Tool - Chintana

4.  Should MANO WG wiki be updated for ETSI updates -Release 2 presented by (Pierre Lynch)

April 12th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #26 summary

1.Agenda Bashing

2. Code Review report of vIMS testing in D1.0 using FUNCTEST - Orchetsra (openbaton as Orchestrator instantiating and testing   openims) - Moderated by Prakash , Giuseppe & Arturt to validate. - Code Review postponed, Dovetail qualification criterion for vIMS VNF testing is immature to qualify for testing at this stage, Bryan suggests focus on VNF Onboarding & Production Deployment for E-Release refer VNF Onboarding

3. TOSCA for MANO scenario descriptor file from last week discussions - Bryan/ Narinder/ Uli - Await inputs -Postponed

4. mano - testing use cases for various vnf types - Arturt - (What will be new adds besides SBC?, vOLTE?, vEPC?) - can be closed

5. Considerations for lower complexity  SBC use case based (define the use case first) on 3-node vHello Tacker in Modles Project (Bryan/Aimee/Prakash) - First Code Review of 3Node v Hello Tacker - BP + Start - Bryan suggested look at VES diagram and may be define the narrow use case for SBC using that for one external, one tenant facing and edge proxy using SBC to reuse the VES model. Plus for SBC as long as we donot modify can we get OK form LF for using this, route through Ray Pike.(Prakash to follow up)

6. Anything new from ONS summit at Santa Clara (April 3-6) on ONAP? - Inputs from Attendees any? - First presentation of ONAP occured at ONS. There is a meet for select ONAP team in coming weeks and we will hear more on that later.

7. Any followup  from MANO WG teams like WSO2 API  - Chintana (Updates for new LF API Tool ?) Chinta reported that they can import any API including Swagger and will review API tools next week as follow up.

8. Events  Plugfest (Paris),  OpenStack Boston + OPNFV Beijing Summit - Any updates? - Plugfest access request must be sent to morganDOTrichommeATorangeDOTcom

9. Should MANO WG wiki be updated for ETSI updates -Release 2 presented by (Pierre Lynch) -deferred


 

April 5th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #25 Summary

1.Agenda Bashing

2. Q1 MANO WG report - Prakash

3. TOSCA for MANO scenario descriptor file from last week discussions - Bryan/ Narinder/ Uli

4. testing vIMS feedback from test WG - Bryan/Prakash on review in Test WG, Orchestra (Giuseppe), Yingjun Li(Opera),

5. Considerations for lower complexity  SBC use case based on 3-node vHello Tacker in Modles Project (Bryan/Aimee)

6. Anything new from ONS summit at Santa Clara (April 3-6) on ONAP?

7. Any followup  from MANO WG teams like WSO2 API and  for Events  Plugfest (Paris),  OpenStack Boston + OPNFV Beijing Summit

8. mano testing use cases for various vnf types - Arturt

Summary of meeting#25 

March 29nd, Wednesday : 14.00 UTC /7.00 PDT  Meeting #24 summary

1.Agenda Bashing

2. Discussions & Actions on Process aspects and API for VNF vIMS Build / Test in  FUNCTET and its value for Release  E. .

3.  Use case discussions  -  vOLTE, can we simplify this?

Summary of meeting

March 22nd, Wednesday : 14.00 UTC /7.00 PDT  Meeting #23 and  Summary

1.Agenda Bashing

2. Discussions & Actions on  MANO Build Options in Release-E - Giuseppe , Yingjun , Arturt, Ulas

3. Streamlining VNF Onboarding & LCM Release-E  and Action plans - Giuseppe , Yingjun , Narinder, Sameer

4. Time permitting Use case discussions -   from Models Test and Samplevnf, vOLTE (Opera request) - deferred for Open-O and Models co-ordination.(Pakash, Yingjun Li, Bryan)

5. Followups on items from last week.

Summary of meeting:

NFVO and VNFM both are external components and OPNFV build gets these files as software artifacts, bundles or as images (VM or containers)

In release D 1.0 the Opera chose to use an openo scenario and Compass Installer and Orchestra uses direct Orchestration from FUNCTEST.

We will have more details once D1.0 is released as what makes in D1.0 and what goes to D2.0

March 15th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #22 and summary

1.Agenda Bashing

     Refer log of Tech meeting last week  MANO WG update - http://ircbot.wl.linuxfoundation.org/meetings/opnfv-meeting/2017/opnfv-meeting.2017-03-09-14.01.html

    Discussed E-Release schedules  and listed that Slides for E-Release planning in 5 below.

2. Review action items from last meeting and progress on  - deferred

     AI101/102/103 - Scenario, MANO Components , External vs. Internal Build  - PTL Opera/Orchestra

     AI201/202 - VNF On-boarding Flow + Use case descriptors (vEPC/vIMS) - PTL Models/Domino - Prakash

3. New timings for Models+Domino for inputs to MANO? - Informed

4. Review MANO WG plans for OpenStack Boston +  MANO WG Survey Questions? -Deferred

5. Discussions on work items identified for Q1 not able to cover from last 2 meetings  if time permits - Discussed as part of slides below

Slides for E-Release Planing

Summary

Delayed start by 10 minutes on GTM because of connection call start issues.

Yingjun Li - Opera  may need extension for Intent to participate in E deferred because of ONAP (OPEN-O & ECOMP harmonization)

Atrurt - MANO integration need 3 pieces: NFVI, MANO and VNF involvement plus sync with Functest

Debyang/Sameer -  Presented updates for vCPE/vIMS based on OAI and believe they can help test and Demo with Orchestra through participation with MANO projects

Giuspee - We need to clarify process flow of Build and deploy over Scenario(s)

Prakash - Agreed to start a thread based on Giuseppe request summarizing the discussions here in

March 8th, Wednesday : 15.00 UTC /7.00 PST  Meeting #21 and Summary

1.Agenda Bashing

2.MANO WG Survey (TBD) / MANO WG Feb report

3. Models/VNF Graph additions from Models/Domino call from this week for use cases through Models and possible SampleVNFs

4. New timings going back to 14.00 UTC ( 7 AM PDT, 9 PM EU, Asia India  IST 7.30 PM / China CST 11 PM)

5. Discussions on work items identified for Q1 not able to cover from last week.- 

Refer log of mano-wg meeting 

http://ircbot.wl.linuxfoundation.org/meetings/opnfv-meeting/2017/opnfv-meeting.2017-03-09-14.01.html

March 1st, Wednesday : 15.00 UTC /7.00 PST  Meeting #20 & Summary

  1. Agenda Bashing
  2.  JIRA for consolidating E-Release Priority : option 1 - Have separate one for MANO WG option 2- Use the ones by respective Projects to refer to - For now option 2 will suffice till we have clarity on ONAP(ECOMP+OPEN-O Harmonization to use Action Item AI)
  3. ONAP and its impact on MNAO WG for OPERA (OPEN-O/ECOMP) - ONAP slide 1.0
  4. Final Agreements for Release E by both Orchestra , Opera and other participating projects + Upstream

Agreements Approval by MANO WG team

Consolidated E-Release Work Items from MANO WG

MANO Projects in E release will focus on test tools for integration and testing MANO stack  from Meeting #18

 Agreed consistency across MANO stacks we mean "Consistency of end users experience" from Meeting #18

Work Items and action items to follow up based on Priorities: ( we will discuss this next week in details)

Priority 1 is Integrating MANO Stack Components over OPNFV VIM from Meeting #19 2017

AI-101 What Scenario MANO Project / General Scenario Reveiew will use in E?  Create  /  Generic or    Specific :Ideally would be use Generic or None. Read Document for CI - Section 7 MANO Scenarios.

Note Both top-down and bottom-up approach must meet  generic OPNFV VIM scenario and some inputs from EUAG for vCPE and NaaS use case for scaling is worth noting.

Should MANO Projects consider : Parent-Child Scenario?

Project::Scenario/Installer/AAA

CI evolution: Boxing Installers into VM on Jumpservers, Should we box NFVO and VNFM into VM on Jumpserver or Compute nodes?

Pharos Specification Update for MANO: Should MANO WG update Pharos Specification last update was update was for Colorado (PHAROS-144)

Copper: none

Domino: os-nosdn-nofeature-ha /JOID / keystone v2 / post-install

Models:

Opera:: os-openo-nofaeture-ha/ compass & JOID/ keystone v3.

Orchestra: not defined

Parser:

AI-102: How do you plan to add your NFVO/VNF or Feature components : Project: Components/Feature review parent/child with os-mano-vxlan-ha as parent and other like opera or orchestra deriving from it as child till difference are settled or we can drop scenario totally.

AI-103: Will you use upstream Build and just provide artifact on a path or URI or use Internal CI evolution to dynamically build and test over vPODs or PODs in OPNFV Community? Describe your MANO Project preference and flow chart for evaluating processes or refer your code.

Priority 2  is considerations for agreement on VNF On-boarding from Meeting #19 2017

AI-201: Describe what is your function flow for on-boarding  (be simple Hello world of vIMS VNF) - We have common ways on on-boarding with some variations  and a common flow or metadata to use?

AI-202: Do you use VNFD(with NSD) or just a simple Blueprint (BP) or TOSCA node template (review  vnf2 yaml file)

Priority 3 is Continuing on Reviewing Best Practices for VNF On-boarding & LCM from Meeting #19 2017 plus EUAG reference

AI-301:  Document or point to vIMS On-boarding procedure in upstream for a) OPEN-O b) OpenBaton c) ONAP/ECOMP

Priority 4 Use inputs from OSM for VNFD and NSD challenges. from Meeting #15 -2016

AI-401: Document and present at MANO WG what we need to l;earn from OSM?

Volunteers who can work on AI on Priority 1 & 2 first.

 

Feb 23: After last meeting of MANO on 22nd a new LF project called ONAP has emerged merging OPEN-O + ECOMP efforts and we cover it on March 1st meeting. See above.

Feb 22, Wednesday : 15.00 UTC /7.00 PST  Meeting #19 and summary

  1. Agenda Bashing
  2. Action Items for last meeting and closures for this month for E-Release Content
    1. Action Item 1 (AI1) of last meeting see Wiki Page for MANO-Best Practices  added besides and needs priority for E agreed here by all Participating projects.

      (Giuseppe - OpenBaton) offers Packages for VNFM for gVNFM as well Juju. Docker based packaging is also being considered beyond OpenBaton 3.0.

      Bootstrap for NFVO+VNFM will be from OpenBaton and Integration with OPNFV VIM will be using JOID Installer + Scenario (to be introduced in E).

       (Yingjun - Opera)  Scenario for OPEN-O os-openo-nofaeture-ha with compass & JOID with keystone v3.

      AGREED: Priority 1 is Integrating MANO Stack Components over OPNFV VIM (Test area 1 will address this in Dovetail through  tests to be proposed for MANO and also the MANO Component Configuration  to expand scope over OPNFV VIM Scenarios.)

    2. Considerations for agreement on VNF On-boarding (Priority 2)

      Refer to VNF OnBoarding and within that End User Advisory linkReview item 4,5,6 from last meeting and Interns 5 priority for VNF Catalog  
      https://lists.opnfv.org/pipermail/opnfv-tech-discuss/2017-February/014950.html - Bryan
      + Based on email reply Prakash
      VNF should establish its name and context as an Application or Component with a path to Application/Component Catalog(Cat.3/Repository). This assumes that hierarchy of Catalogues/Repos is Cat.5/Repo5 (Marketplace), Cat.4/Repo4 -Service Provider Service Catalog (This is where you on-board NS/NSD), Cat.3/Repo3 - Application/Component Catalog (This is where you on-board your VNF/VNFD/Packages etc), (Cat2./Repo2) -SDN Controller Catalog. Cat1/repo1 is VIM/Glace Catalog. Catalog Cat0/repo0 will be instance Catalog. e.g.. HEAT.    
      (Giuseppe) Define first the metadata and content like using Blueprint etc before a formrat can be made or agreed.
      How you create package is through SDK which is being proposed by both OpenBaton and Open-O 
    3. Plugtest and inputs for MANO from Participants to see if anything can be used for E-Release content - Since there was a nondisclosure agreements by participants this item is closed for now.

  3. Code review for VNF testing - Completed  common testing for vIMS (Orchestra to use Orchestration and Opera to use Compass Installer for testing vIMS as VNF use case). Added request for Onbord enabled flag and externalization and add to Service Catalog of enabled, Package name/type and descriptor explicitly through extended class to baseVNF class as per new files for Opera and Orchestra.
    What will be selection of use case and test areas/tests which we can drive MANO inputs for D release of Dovetail (vIMS - Test Area 2 - suggest adding vEPC to existing vIMS and improvise to add End to End Packet Data Besides Voice services- needs input from Opera and Orchestra.

    Continuing on Reviewing Best Practices for VNF On-boarding & LCM (Priority 3)
    a. MANO SB APIs Or-Vi is most mature at this time to consider for E-Release
    b. Close loop monitoring is a possible candidate for OPEN-O with Doctor Project (Giuseppe for OpenBaton in E-Release)
    c. Go beyond to adding Policy for VNFD as use of Simple TOSCA YAML file which is current norm for most MANO projects. 

  4. OSM & ETSI NFV updates - Refer [ RP ] in Meeting #15 -2016.

Feb 15, Wednesday : 15.00 UTC /7.00 PST  Meeting #18 continued from last week and summary

  1. Agenda Bashing
     
  2.  Report from Upstream  integration - OPEN-O /Opera - Yingjun Li
  3. Report form OPNFV participating project PTLs  for better Orchestration & Automation i E-Release.
    1. Models Meeting#24:  - Models to focus on completing Blueprints of VNFDs support in E-Release. Visibility to Upstream limited, high level abstractions like in OPEN-O not visible to OPNFV - Info Models support from ARIA, OPNFV, OpenStack, Kubernetes etc. At test level will continue  to shift from Stable to  Trunk - Continuous Development on upstream Trunks. Primarily OpenStack and beyond like OPen-o, ONOS, ODL, OpenContrail etc.
    2. CopperGot Congress support in Installers but  ODL GBP lack of contributions.  Continue to support E-Relase but may Sunset in D time frame, and move everything upstream, Remaining FUNCTEST etc be in upstream, deployment should possibly be  installer agnostic.
    3. Parser: Await inputs from OPNFV Parser, Inputs from ARIA.
    4. Domino: Evaluating Integration for NFVO Registration with OEPEN-O/GSO, integration with OPEN-O GSO using TOSCA and policy , considering multi-parser etc.
    5. JOID(LXD): Workloads of K8 for LXD in E. link https://bugs.launchpad.net/charms/+source/percona-cluster/+bug/1663245
    6. VES :  Analytics in closed loop policies , Barometer has few and therefore Integrate and test using VEs & Barometer. Another possibility is Redhat ManagedIQ. Use Cases like Fault, scale, push, pull, performance requirements
  4. Review and Guidance for MANO based on newly specified OPNFV Architecture - OPNFV Architecture for Danube release - New from TSC call 1/31/2016

Feb 8, Wednesday : 15.00 UTC /7.00 PST  Meeting #18 and Summary

  1.  Agenda bashing
  2. Follow-up on Action Items and  from last month and finalizing E-Release Goals and agreements
    1. Goals for E should be to release test tools for integration and testing MANO stack -Bryan
       

      E-Release Goals and Contents

       

      E release must focus on test tools for integration and testing MANO stack

       

      Agreed that when we mean Consistency across MANO stacks we mean "Consistency of end users experience"

       

      This means each MANO stack may follow different formats and methods but process as it appears to an end user is consistent.

       

      AI 1: Compile VNF Formats , Onboarding process, Catalogs and LCM for few stacks to understand the differences and try align for consistent end user experience.

      (Prakash will initiate a Wiki page and Bryan and others will edit and update it for next meeting to go over rest by emails)
    2. Release E discussions: using keystone Auth v3 in Release E - Moved from D to E -D2E - Guidance is OK and its no binding and will depend on the progress of each MANO Stack and its interaction with installer.and OpenStack.
    3. Review and Guidance for MANO based on newly specified OPNFV Architecture - OPNFV Architecture for Danube release - New from TSC call 1/31/2016 -This could not be taken up deferred for future meetings.
    4. Report from upstream (OpenBaton, Open-O, openECOMP, JuJu, Tacker, OSM/Plug-test) and its impact on MANO WG - Details of Inputs from MANO Projects refer below Details of Discussions in Meeting#18

  3. How to arrive at best practices for NS and Cloud Native VNF Package Formats and run time, - Deferred to next meeting- Add to AI 1 for next meeting to review on Wiki
    1.  Generating YAML(TOSCA/YANG)  refer link  Models VNF-Packaging and Riftt.io , TOSCA file inside CSAR zip file
    2.  Linux Package-Management formats , OpenSUSE Package Manager
  4. Follow-up on Action Items and  from last month and finalizing E-Release Goals and agreements (must be before March 27 as per EUAG) - Add to AI 1 for next meeting to review on Wiki
    1. NS/VNF Package Standards adapted ( TOSCA/ CSAR (OPen-O)?, TOSCA / rift.io Juju YAML (OpenBaon?), TOSCA/Rift.io (OSM), ECOMP ? - AI who?
    2. VNF On-boarding Process across MANO stacks (to be consistent? how?) - AI (who?)
    3. Keystone v3 - (Installers JOID/Junju in E?)
    4. Use case continue with simple Hello world + vIMS or add one more in E?
  5.  Plugtest and inputs for MANO from Participants. - Deferred to next meeting

    Details of Discussions in Meeting#18
    OpenBaton : Guiseppe on Architecture comparison
    We have one done as it exists and evaluated in January 2017: https://wiki.opnfv.org/display/mano/Architecture
    OpenBaton has  Project dependency - Need help  from JOID and Mile Stone support (MS?). Feature projects don't have all Milestones to meet
    David McBride is working in Release Management team to simplify the Milestone process to be separated for Specification, Feature Projects, Installers etc.

    Being a Feature Project and working through FUNCTEST in D Release the support is limited to FUNCTEST of vIMS use case
    JujU and JOID are not same, JujU is a GVNFM and Canonical does work with OpenBaton.
    Issue is more of Integration and testing with FUNCTEST whic uses Swagger APIs, and Morgan has been helping OpenBaton.
    The FUCTEST team has reduced pain point for OpenBaton by providing Orchestration option in FUNCTEST for VNF (vIMS)
    Similarly FUNCTEST has allowed Open-O to use Installer option through JOID and Compass and for Compass Open-O is using Keystone V3 in its scenario.
    Based on D Release experience OpenBaton team like to start integration from beginning of E-Release
    OpenBaton will target  Doctor for E version, Needs documents ( Colorado is here http://artifacts.opnfv.org/doctor/review/23963/releasenotes/index.html)
    For D-release please check with PTL Ryta Mibu ( https://wiki.opnfv.org/display/meetings/Doctor )

    JOID/JuJu Canonica:l Artur/Narinder mentioned that Charm for Orchestra is ready and should be able work throug JOID too and OpenBaton can work with JOID team.
    Testing code for vIMS is getting ready, link https://gerrit.opnfv.org/gerrit/#/c/26769/
    Artur from Canonical like to see in Release E, use of JuJu and Charms in  Open-O, OSM, OpenBaton part of Functest  Scenarios etc.

    Cross Project Dependency
    Can we invite Doctor PTL or Committer to present in MANO WG as how to integrate with different MANO stacks? -follow up through email for now.

    For Models what can be used  Giuseppe and Bryan will work out using Python once OpenBaton has some codes to use in Gerrit in Orchestra project.

    Discussion about scenarios and features...
    architecture - management models NB
    Potentially there will be test that will integrate into ECOPM , are working on VES
    How MANO projects address how open-loop and close-loop control and telemetry , LCM , using Celometer/Heat, Redhat has Insight and Satellite 6  , the other is Barometer in OPNFV etc
    Add to AI1 above : Service creation., onboarding, overall orchestration, Application LCM, Inventory system, Analytic (Close loop system), Plugin like Policy, Security and Resource Orchestration etc.
    Action Item for Prkash to create a wiki table..Prakash and Bryan will support it to enlist items in AI 1 here and above.

    Sridhar K.N. Rao /Solution Architect wanted to know if test tools in Release E refer to Testing Projects OR open Source tets tools like REST API for VNFM , testing Framework like Dredd?
    https://dredd.readthedocs.io/en/latest/ & https://apiblueprint.org/tools.html - May be ask for a presentation from Redhat or Sridhar to present this if MANO WG consider this as a value for E-Release
    Bryan Sullivan answered : The point being made in that note was the we should be working tactically to setup test environments for MANO stack testing now and thru the E release.
    I do not believe that in the next 6 months we will have the ability to drive convergence in any real sense, except thru use case testing with existing open source MANO solutions.
    In the Models project we are working to establish a pattern for such tests, focused on assessing the set of blueprint features supported by various MANO stack components.
    There is a similar effort underway as mentioned by Giuseppe in the Functest project.
    I had recommended earlier that test approaches to MANO functionality should be in the scope of the MANO WG, and that Functest should simply incorporate what the MANO WG develops as a framework.

    But in the meantime progress follows momentum, and that momentum is in the Models and Functest projects.

    Meanwhile we did not get inputs from Yingjun Li from OPEN-O /Domino/ Parser and like to schedule inputs at  meeting#19 before we conclude on E-Release updates
    Time lines: By Feb 23 to bring it to Tech Meeting before getting to TSC by Feb 28

Jan  25, Wednesday : 15.00 UTC /7.00 PST  Meeting 17 and Sumary

  1.  Agenda bashing
  2. Review of last meeting (#15) and actions
    1. Priority for D - completed and accepted
    2. Review of VNF on-boarding - Is anyone Interested in monitoring Intern for VNF On-boarding - No inputs yet
    3. Code review for VNF testing - Completed  common testing for vIMS (Orchestra to use Orchestration and Opera to use Compass Installer for testing vIMS as VNF use case) - To review feedback next meeting along with plugtest.
    4. Scenario conclusion - Orchestra , Opera, Domino , Model , Parser, - Do we have a path to no scenario? Need to fix a simple framework and refer  Models -see vHello_Tacker.sh and Danube Scenarios.
    5. Priority Release E (Move left out D to E + for consistency)- Goals for E should be to release test tools for integration ans testing MANO stack
    6. ABOT framewrok for BDD has impact beyond MANO and hence moved the discussions to Tech Meeting - Completed Transfer to Weekly Tech for Thursday.
  3. Architecture comparison and LCM.
  4. OSM & ETSI NFV updates - Refer [ RP ] in Meeting 15 -2016 -Deferred to next meeting after Plug-test
  5. Report from upstream (OpenBaton, Open-O, openECOMP, JuJu, Tacker, OSM/Plug-test) and its impact on MANO WG -Deferred to next meeting
  6. MANO related tests for Dovetail beyond C for D based on what we have and who will volunteer to get this moving? - Dovetail uses Functest but we should invite them to present their tool for benefit of MANO projects or to Tech meeting so we understand what they did in C to see how it will move up the stack in D and help MANO testing in future.
  7. Continuing on Reviewing Best Practices for VNF On-boarding & LCM based on current community project status? - Strategy towards tool deriving from Models like Cloudify , Tacker and will need more discussions with participating MANO projects to help in this efforts.
  8.  Actionable Items and Agreement summary recording - A1. Bryan for wiki Architecture page move, A2. Prakash for Release E update & Getting Dovetail Test tool presentation in MANO or Weekly Tech meeting.

Jan 11, Wednesday : 15.00 UTC /7.00 PST  Meeting 16 & Summary

  1. Agenda bashing
  2. Review of last Quarter (2016) work - Bryan Sullivan on what we did and what we missed (Set Priority) - Proposed VNF On-boarding and Pilots for testing and satisfied with progress
  3. Plan priority for D & E releases of OPNFV ( Review request for VNF OnBoarding for vIMS) (Code-Review or testing Vnf) - Agreed on D-Release Priority write up reviewed link and plan for E future meetings, Code Review lest to participating projects to comment on abstractions and ability to adapt by OPEN-O, OpenBaton and other teams. Some comments from Serge Manning (Sprint) on Inter-operability and Bryan indicated that goal is to move towards consistency and portability for providers & end users for VNF LCM etc. The AAA and Keystone V3 was discussed and David McBride Release manager pointed to link for same.
  4.  Common Test for vIMS (clearwater) - This was used as starting point for testing complex VNF like vIMS and code is being adapted for testing as in 3.
  5. Opera & OpenBaton Review (see links below) - Refer [ YC ] and [ GC ]  in Meeting 15 -2016 - Both Giuseppe and Yingjun Li represented Orchestra and Opera respectively and indicated continued participation in testing vnfm for D release. YC to allocate a code reviewer for testing vnf link form OPEN-O team, Giuseppe has reviewed and looks OK but needs more details to be included from the points listed in Meeting 15 content below.
  6. JuJu VNFM and mano Scenario support for JOID - Refer [AT ] in Meeting 15 -2016 - Artur and Narinder indicated OAI as EuruoCom initiative and also pointed to Deb from Rebaca.
  7. ABOT Framework feedback - Refer [ DC ] in Meeting 15 -2016 - Due to lack of time and also nature of DBB and testing it is considered this be presented at Tech meeting and we will check with Bin Hu if he can help get this session instead of MANO as it covers wider involvement of community to assess this for OPNFV.
  8. Architecture comparison and LCM - Deferred to future meetings.
  9. OSM & ETSI NFV updates - Refer [ RP ] in Meeting 15 -2016 - Deferred to future meetings.
  10. Any action plan for Q1 2017.- Just send the link of Release - D priority to community and see if this will suffice for Q1/2017.

Nov  30: Dec 7,14  2016 (Q4) Meeting 15 (Hackfest + Ad Hoc):

[ YL]  Yingjun Li (Opera) provided details of Hackfestprep and delivery

https://wiki.opnfv.org/display/meetings/Opera

https://wiki.opnfv.org/display/PROJ/OPNFV+OPEN-O+Integration

https://wiki.opnfv.org/display/PROJ/OPERA+Function+Test

[ BS ] Bryan Sullivan's Input from Hackfest for use of Secnario and Installations: refer inside to POD/Genesis support MANO vPODs

https://etherpad.opnfv.org/p/infra_dec_hackfest

[ GC ] Giuseppe Carella (Orchestra) provided details for OPNFV MANO Integration.

In general, in Orchestra we would be interested in having some integration scenarios which could be useful for covering test cases for MANO frameworks. We have already an upstream project [1] doing something similarto VNF Onboarding functest, and our initial plan was to integrate this project within OPNFV functest, as already mentioned earlier to Jose. With the approach taken in your vnf onboarding abstraction I see that this integration won’t be really easily doable. Main reason is that our integration-test framework already splits the test cases in different steps, while you cover at the moment mainly two steps (deploy VNF and test VNF)

* registration of a VIM instance 

* on boarding a NSD

* deployment a NSR using the on boarded NSD from previous step

* testing that the NS is really deployed (optional in case of negative testing)

* deletion of the NSR

* deletion of the NSD

* unregistration of the VIM

 

With our framework, it is possible to define a scenario (and the steps which needs to be executed) directly using a configuration file. An example of a scenario can be found here [2], deploying an iperf server and client network service and testing that the client connected effectively with the server (therefore testing that dependencies were correctly configured between VNFs). 

 

Although I see that your approach can be further extended to include those additional steps, I would suggest not to have them as part of VNF testing, rather starting creating a new set of test cases which can be part of a MANO group (or maybe including them in the feature tests group?). Nevertheless, coding the test steps in a specific programming language will also limit the integration of existing external frameworks (like the one we already have) MANO providers may already have implemented. 

 

In summary, talking about the Orchestra project, we will provide soon the code for testing the deployment of a VNF (vIMS based on OpenIMSCore) implementing the abstract class you provided and following the approach taken for cloudify. However, as we assume that in the future the Orchestra project will be directly integrated with at least one installer, we would like to include some more comprehensive use cases that can execute several (positive and negative) test cases, possibly using already existing frameworks (like the one I mentioned). Would that be feasible within functest? 

[ 1 ] https://github.com/openbaton/integration-tests

[ 2 ] https://github.com/openbaton/integration-tests/blob/master/src/main/resources/integration-test-scenarios/scenario-real-iperf.ini

[AT ]Canonical's and OAI input for VNFM

Canonical and OAI inputs to OPNFV MANO for VNFM

[ DC ] Debyang Chaudhari of Rebaca Automation on ABOT BDD test framework

ABOT BDD Framework

[ RP ] Ramchandran Parkash on following on VNFD and NSD information Model Challenges in ETSI NFV

VNFD Challenges

·         VNFD Connection Points

·         VNFD Deployment Flavour & associated changes in VDU topology

·         Lifecycle management events and scripts referenced in the VNFD

·         Enhanced Platform Awareness Attributes

NSD Challenges

·         Deployment flavour complexity

·         Nested service ambiguities

·         Local Affinity Rules vs Affinity Groups

·         VNFFGD and the lack of relationship with deployment flavours

·         Lifecycle management events and scripts references in the NSD


Changed to weekly meeting due to MANO activities in Q4 2016.

Agenda of meeting November 30, 2016  15.00 UTC /7.00 PST Meeting #14 and Summary

  1. Review of last week work and Agenda bashing

      2. The  FUNCTEST-353 is more comprehensive than  focus on API testing through common wrappers. Therefore  4 mores tasks are created under Functest to address that and they are

        a.Understand the architecture ( Inbuilt as part of OPNFV, Install as part of Orchestration the NFVO, Install and and test ).

        b. Adding NFVO to installer

        c. How to test or-vi - FUNCTEST-614,

        d. How to test or-vnfm -   FUNCTEST-615

       Yingjun/Giuseppe please ask your teams to  compare with https://git.opnfv.org/functest/tree/testcases/vnf/vIMS?h=stable/colorado this code with Java codes presented last week and give feedback.

     3. Action Artur to plan for getting Rebaca Automation team to present on how they use Juju with NFVOs - ABOT Presented and reviews still in progress

     4. How would the results of collaboration through MANO WG reflect in D-release of OPNFV. - A document template  to be proposed and  reviewed to see what should the best practices to include in the D-release based on MANO WG participation by Opera, Orchestra and associated VNFM.

     5. Discussions on what would be inputs from MANO_WG to OPNFV D & E Release. Wiki Page to update after discussions in WG. - Bryan to update this for Release D with minimum and Prakash to review all the participating projects and provide a report this meeting on 30th Nov.

     6. Any  Updates on UNH Plug-fest

Agenda of meeting November 23, 2016  15.00 UTC /7.00 PST Meeting #13 and summary

  1. Review of last week work and Agenda bashing

    Need FUNCTEST-237 replaced by one that Opera/Orchestra will commonly use - YingjunLi/Giuseppe)

    Giuseppe went over the codes listed for wrapper calls for or-vi and or-vnfm and explained how it uses Java Public calls to invoke VIMDriver which can be a Python Client or API calls to OpenStack VIM.

    https://github.com/openbaton/openbaton-libs/blob/master/vim-int/src/main/java/org/openbaton/nfvo/vim_interfaces/vim/Vim.java

    The code below is similar code for or-vnfm for VNF lifecycle management calling the rabbitMQ and can be aligned for calls to JuJu or tacker with some common agreements to be discussed next week.

    baton/openbaton-libs/blob/master/vnfm-sdk/src/main/java/org/openbaton/common/vnfm_sdk/interfaces/VNFLifecycleManagement.java

    For each of the  projects OpenBaton/Orchestra, Open-O/Opera please detail the items A,B,C and D below (Bryan to reword and fix the same below)
    A. Compile workflow of LCM  of a VNF.

    B. Compile  the  MANO  NFVO components/Blocks  for OPNFV MANO stack Integration. - Action Item on Prakash

    C. Define and describe the  role of VNFM (Juju, Tacker) if these are used in B as options. - Action Artur to plan for getting Rebeca team to present on how they use Juju with NFVOs)

    D. Compile Sequence Diagram for LCM of VNF through the components/Blocks (NFVO, VIMDriver, VNFM,...) - Giuseppe suggested this to be able to co-relate with codes links listed above.

  2. Where is the test plan and how will the projects share the testing codes? - Proposal to be discussed next week with inputs for Giuseppe of Orchestra (OpenBaton team). -Deferred to item 2/Nov30
  3. How would the results of collaboration through MANO WG reflect in D-release of OPNFV. - A template to be reviewed to see what should the best practices to include in the D-release based on MANO WG participation by Opera, Orchestra and associated VNFM. -- Deferred to item 3/Nov30
  4. Discussions on what would be inputs from MANO_WG to OPNFV D & E Release. Wiki Page to update after discussions in WG. - Bryan to update this for Release D with minimum and Prakash to review all the participating projects and provide a report next meeting on 30th Nov. 
  5. UNH Plug-fest for cross-project co-ordination? - Update from OPEN-O/Opera team - This Bryan is working with JuJu team to see if any VNF cam be demonstrated at UNH. Plus to avoid any sign-up issues the Plug-fest and Hack-fest discussions on MNAO are separated out.

Agenda of meeting November 16, 2016  14.00 UTC /7.00 PDT Meeting #12 and Summary

  1. Summary of last meeting and Agenda bashing - No additions
  2. Compare Clearwater vIMS vs. OpenIMS- (Artur Tyloc - details on OpenIMS and JuJu orchestration for the same)
    Bryan asked what comparison means and was agreed that we work to find on-boarding package of vIMS and workflow for LCM of vIMS.
  3. Review  or propose TOSCA template for vIMS and OpenIMS for On-Boarding and compare with ( NFV (VNFD, VLD, VNNFGD, PNFD) - (Yingjun Li) - the  https://github.com/Orange-OpenSource/opnfv-cloudify-clearwater (openstack-blueprint.yaml) has details. His team is working to understand and use it for now for UNH Plugfest.
  4. What VNF  package should  contain besides TOSCA descriptor ( for CW vIMS / OpenIMS) - OPEN-O & OpenBaton for their NFVO & VNFM choice they make (Opera/Orchestra teams) - snap from Canonical http://snapcraft.io/ was discussed and other one we had looked earlier was CSAR
  5. Service Invocation and Termination API versioning and flows for VNF (vIMS/OpenIMS) use case.(General Discussions) - Agreed that we must include verisioning while using API for LCM of VNF.
  6. Inputs from OpenBaton for IFA5,6,7 Interface functional calls. (Giuseppe) - Lorenz pointed to various nfvo/vnfm scripts in https://github.com/openbaton
  7. Progress on Models  based on Peter Lees inputs using Django & YANG. How will that apply to vIMS/OpenMANO? (Bryan)
    Bryan added link  for comparison - Comparison, references , testing


Agenda of meeting November 9, 2016  14.00 UTC /7.00 PDT Meeting #11 and summary

  1. What should we document for MANO best practice for OPEN-O and OpenBaton integration through Opera and Orchestra.
    1. What use case to start with - - Should vIMS be the first use case?
      vIMS is agreed use case to document as first one for MANO integration
      Besides Clearwater vIMS Blueprint in OPNFV,  Arthur from Canonical suggested using OpenIMS and JuJu can handle both.
    2. VNF On-Boarding - TOSCA Node Template <similar to> NFV (VNFD, VLD, VNNFGD, PNFD) - Onboarding
      OpenBaton / Orchestra indicated that they like to see VNF Package to include VNFD along with image name and script to instantiate the VNF. So did Yingjun Li PTL Opera chimed and thus there is agreement on this from both projects. The issue of formats to use include CSAR & JSON and run times both Python and Java are being considered. Thus there will be attempt by both teams to define them along with VNFM's they plan to use like JujU and Tacker.

      The VNF package, and data model / representations for NSD and VNFD, and how/why those vary across MANO projects

      b1.       OpenBaton: JSON for VNFDNSD; TOSCA for NSD

      b2.       Open-O: …Needs input from Opera team

    3. Service Invocation and Deletion - TOSCA Service  template = NFV NSD, How do we handle SO & RO for this?
      This will follow NFVO flow to VNFM asking to allocate resource and if it does not that capability then NFVO will reroute the request to VIM. On success of Resource reservation (Quota) etc. the NFVO will ask VNFM to instantiate the service for the deployment completion as first target. Life cycle management will follow that based on agreed hooks to be defined as common between the Opera , Orchestra, Juju and Tacker teams. A proposal will be reviewed next week.
    4. API versioning - which API?
      The API versioning is of abstractions or wrappers that is being proposed by OpenBaton and will be reviewed by other MANO participants to agree on a common wrapper approach for IFA 5, 6,7 and OPNFV VIM and VNFM drivers from NFVO.
  2. At what point should upstream integration project  consider OPNFV Models topology design and templates to choose  and use. - This will be discussed at next Models call with Bryan Sullivan.
  3. What should be test strategy (refer  item 3 Nov 2 meeting) - Await FUNCTEST-237 response from Morgan and Action for MANO teams to create new FUNCTEST JIRA ticket based on vIMS use case.
  4. Where is the test plan and how will the projects share the testing codes? - Proposal to be discussed next week with inputs for Giuseppe of Orchestra (OpenBaton team).
  5. How would the results of collaboration through MANO WG reflect in D-release of OPNFV. - A template to be reviewed to see what should the best practices to include in the D-release based on MANO WG participation by Opera, Orchestra and associated VNFM.
  6. What is reasonable deliverable expected from UNH Plug-fest for cross-project co-ordination? - The Open-O team from China Mobile plans to demo OPEN-O and OpenBaton team indicated that they may not be bale to make it to UNH Plug-fest but plan to Demo their efforts in ETSI Plugtest in February.
    Attendees: irc+GTM

Agenda of meeting November 2, 2016  14.00 UTC /7.00 PDT Meeting #10 and summary

1, Briefing on "State of MANO architecture" and feedback listed below.

MANO TSC-Board report and feedback
State of MANO in OPNFV-tsc-borad-october-28-2016.pdf

2. Cross-Project Co-ordination for Danube and Assess the participation of MANO  activities for UNHV Plugfest Dec 5-8

3. Functest through common test plans for MANO VNFM and NFVO  (Opera, Orchestra, and participating Projects)

Added comments to JIRA ticket. Await response from Morgan if we need a new JIRA ticket or plan to persue with this for both Opera and Orchestra for VNF Blueprint testing.

https://jira.opnfv.org/secure/RapidBoard.jspa?rapidView=69&projectKey=FUNCTEST&view=planning&selectedIssue=FUNCTEST-237

4. Any topics of interests to upstream and Projects in OPNFV.

Updated the ETSI Mano page for NSD, SO/RO with acronyms in alpha order.

https://wiki.opnfv.org/display/mano/ETSI-MANO

Updated new  weekly scheduled meeting Wednesday 14 UTC for MANO WG  at top of this page.

Agenda of meeting October 24, 2016  20.30 UTC /13.30 PDT Meeting #9 and summary

  1. Inputs to TSC for Board "State of MANO architecture (Multi-modal)"
    1. Predictable - VNF Onboarding - Discussions
    2. Exploratory - Integrating MANO components in OPNFV and Hello world  Blue Prints efforts for VNF LCM.
    3. Evolving Architecture - Top Down vs. Bottom up - Prakash
    4. MANO Project Guidance
  2. Review Internal Process
    1. Need change of time (with Daylight Saving changes and to help EU/China participation)
    2. Need web page updates and review.
    3. Stick to one irc channel allocated : #opnfv-mano and help needed for taking notes

Summary:

 Actions : 1.  Prakash  will work with Bryan Sullivan to prepare a Deck to present on "State of MANO in OPNFV", at TSC/Board meeting at Barcelona.

                2. David McBride will get in touch with Orchestra PTL [Giuseppe Carella (Frauhaufer de)] to encourage them to run through the D-release plans for OPNFV integration.

                3. Opera PTL[ Yingjun Li (Huawei)] went through process to adapt, for Integrating NFV-O and VMFM over OPNFV VIM as in item a & b and the edits reflect the changes in those links.

                    Main point to note is exception for having a scenario in OPNFV to catch up on version mismatches for OPEN-O to OPNFV. This scenario may need resources for CI/CD which will be evaluated for Release D.

                    Opera PTL

                4. Kashyap from HP joined on understanding OPNFV approach to on-boarding and due to lack of time we discussed only catalog portion and that resulted in action item for mano wg as a whole..

                    Thus focus must b on what are available VNF Packaging formats and which are widely used and why as part of VNF on-boarding. CSAR link in VFN o-boarding needs to be current from OASIS.

                5. Bryan pointed the requirements brought out by upstream like OPEN-O and vendor Intel who were interested in Market Place and that's added as one item under VNF Boarding and can be edited and expanded by mano wg.

 

Attendees:

Prakash Ramchandran (Futurewei Tech. Inc)

Bryan Sullivan (AT&T)

Lawrence Lamers (VMWare)

Yingjun Li (Huawei)

David McBride (Program Manager -OPNFV/LF)

Kasyap (HPE - NFV)

 

Agenda of meeting October 8, 2016  20.30 UTC /13.30 PDT Meeting #8

Agenda:

Agenda bashing

Briefing on Open Daylight Developer Design Forum between MANO WG and Polestar (Process & work split)- Bryan

Continuing debate on on-boarding flow and processes.

The architectural issues of "catalog" amongst different VNFM, NFVO and possible options to minimize gap

action follow ups...

Summary of last meeting:

1.relationship between Polestar and MANO WG (dneary, 20:36:58)

a.Prakash reported a conversation between Margaret Chiosi, Chris Price and himself (dneary, 20:37:59)

b.Yingjun told us about VNF onboarding during Polestar meeting in ODL Summit in Seattle, but left before the end and did not see the conclusion (dneary, 20:38:44)

c.Bryan has agreed to use Jira for tracking VNF Onboarding task according to Prakash (dneary, 20:39:34)

d.conversation re VNF onboarding - 2 topics including VNF onboarding and VNF packaging (dneary, 20:48:53)

e.3rd aspect is VNF catalog (dneary, 20:49:45)

f.further discussion on how to include VNFM in scope for OPNFV - do we need VNFM scenarios? There was a suggestion to deploy the VNFM as a virtual application through Functest as an application running on the NFVi (dneary, 21:27:22)

g.Prakash is looking for a PTL for a project to start a VNF Catalog project (dneary, 21:31:59) - Could not confirm from OPEN-O as they were planning to use OPNFV artifact url as  Catalog.

Agenda of meeting Sept  21, 2016  20.30 UTC /13.30 PDT Meeting #7 (pre-ODL summit meeting) and Minutes

Circled through listed Agenda

Discussed Proposals from PowerPoint listed in agenda. 

Role of Polestar (user view/ requirements templates, concept validation, Concept Overlap)

Role of mano-wg ( assess and document best practices working with projects that work in mano stack, also share info with other working groups)

Pointed some comments in mailing about Dovetail separation

Details to be discussed at ODL Summit by Polestar / mnao-wg refer link Meetup at OpenDaylight Developer Design Forum

Questions on Specs and   Blueprints : we need terminology change in specs means OPNFV specs

Blueprint here may mean VNF descriptor, Information/Data Model, NSD , VNFFG we need some term for Blueprint to describe this precisely

 Action : Prakash to send summary though dlist.

Attendees:

Prakash Ramchandran - Futurewei Tech. Inc.

Serge Manning- Sprint

[Goal: Documenting specification of Best-practices for MANO]

Agenda of meeting Sept  7, 2016  20.30 UTC /13.30 PDT Meeting #6

Agenda Bashing

mano-wg-plan-v1.0

(JIRA tickets can be optional for Projects that want tracking)

1. VNF On-boarding scope in Polestar vs. opnfv-mano-wg (user view vs. Practice)

Please subscribe to mano-wg through this link

Models Meetings (refer October 19th/2016 notes of Bryan)

Domino Meeting Notes (refer October 19th/2016 notes of Ulas)

Projects and topics that thread through many of the systems and processes that make up a MANO stack, as above (e.g. business processes that are outside OPNFV/OpenSource scope).

VNF Packaging (Models -Pending) - Compare and contrast  in how JuJu and Tacker  use OpenStack Glance or SDN-C (ODL/ONOS/OC}  {how do they align with IFA011?}

This analysis is underway (or at least in scope) in the Models project, as a test/documentation exercise. The current state of that project is setting up the test environment (Cloudify, Tacker, JuJu, OpenBaton, etc) to assess how MANO-stack projects require/support various VNF Package elements, starting with TOSCA-based blueprints for reference VNFs (which will become part of the OPNFV CI/CD inventory along with vIMS). I welcome the support of anyone that wants to focus on the specific question Prakash mentioned.

2.Mapping between TOSCA and NFV for using VNFD from oasis for VNFM to use (http://docs.oasis-open.org/tosca/tosca-nfv/v1.0/csd03/tosca-nfv-v1.0-csd03.pdf ) - need - Arthur Tyloc & for Tacker (TBD) along with OPEN-O/OpenBaton Participation here

? Review how Service Descriptors from TOSCA match with NFV and if mapped correctly can  Template Distribution and Parsing manage this to deliver to orchestrators   like Heat or any controller that can support Pub/Sub via OPNFV project like Domino/Parser?

Bryan -> VNF Package cataloging/distribution is part of on boarding process.( with other VNF assets). The mapping of elements (e.g. Service Descriptors) is part of the information model – data model analysis that needs to be done; work that so far is still in the SDO realm AFAICT.

3. OPEN-O and use case for vCPE - (Opera)

4. OpenBaton use case vIMS/VNF On-boarding (Orchestra)

5. VES & Monitoring formats and  attributes to collect? (This work is listed in Models working with VES)

6. How do we document analysis besides Wiki page updates, any suggestions?  refer mano-wg-plan-v1.1

Please add any items you like to bring this week above as item 6...etc.

Agenda & Summary of meeting Aug 24,2016 (Toronto) Meeting #5

Includes both Toronto inputs and subsequent meeting same day:

1. Welcome to MANO WG -

Goals: A. Guide and Support for standards to adopt. MANO Stack in OPNFV &  B. Roles to ensure Service Providers and Vendors convergence on gaps to cover leveraging this WG.

2. Agenda bashing and last meetings summary

There was suggestions for improving communication and focus by  Hackfest and summarized here by Prakash.

.a. The link to meeting time and location for MANOP WG must be synched or maintain a single copy and send invites to participants via   [opnfv-tech-discuss] [manowg]

Cross Projects - Tasks that were proposed and discussed and how to collaborate through Projects

Priority is Tagging MANOWG  in OPNFV Project JIRA tickets(ask OPNFV Projects for JIRA ticket in Projects to support MANO WG participation) What will be useful, can it be targeted (Chris Price)

b. VNF On-boarding (Models, Copper, Domino, Parser) - Can we make it Policy driven and if so how? [Larry Lammers -VMWare] Modeling language agreement, Announce you capability use service like Domino to match the capabilities requested by Network for VNF Domain. Policy rule definition to support affinity/anti-affinity rule agreement using common labels

c. Project and MANO WG can look at Monitoring aspects for FCAPS, RAS (Dave Neary - Redhat) Doctor - VIM events, Triage, Integration with Monasca, (alignment across multiple projects - Infra tools, how do we go about formats)

d. Policy rule -format specify rule, type of rule, properties key/value, targets are VNF, VDU, COnnection, VL , VNFFG (CRUD node / server ) Domino parses policy and extracts and serializes the nodes and capabilities in terms of exracted labels - Ulas Kozat - Huawei

e. Fault Data Collection and agent (VES+Doctor) - Dave Neary (Red Hat) Mointtoring Infra , Event, log Management, Telemetry, Application couters/gauges, dashbaord Tools common ones like Nagio, Sensu, snap traps, Analytics engine, Panda/Spark,etcd (requirements - inputs from Operators)

f. Catalog - URI and (Murano, JuJu, Tacker, Opera, OpenBaton) - Narnder Gupta (Canonical) Requirements Collection for info for Orchestration and what is the commonality across them that OPNFV can suggest as best practice Orchestrator goes through OPNFV (have VIM, SFC,SD drivers), Can OPNFV help formalize that

g. MANO projects: Currently participating (OPNFV+Upsream)...OPEN-O (Opera),OpenBaton(Orchestra), Models, VES, Domino, JuJu, Tacker

 

4. What is the incentive for other MANO related (Projects+Upstream) : Doctor, OSM,ECOMP,...

This is covered in Summay above and no representation from Docor or OSM yet in MANO WG.

5. Cross-project Co-ordination (goes with Models Meeting Monday even weeks monthly mornings/evenings) -Summary

The key will be to get MANOWG tags

6. VNF On-boarding (goes with Models Meeting Monday even weeks monthly mornings/evenings) -Summary

This will be the first effort for MANO WG to debate and provide best practices as guidelines through  OPNFV participating cross projects.

Attendees:

Prakash Ramchandran - Futurewei Tech. Inc.

Larry Lammers - VMware

Narinder Gupta - Canonical

Ulas Kozat - Huawei

Raghavendrachari Kamsali - HPE

7. Catalog Interfaces - Demo

8. MANO WIki Page updates.based on inputs from participants.

08/10/2016 Meeting #4

Summary of  meeting (same as last meetings agenda)

Establishing an irc with meetbot #opnfv-mano - Completed

Wiki page review and update - Continues

Other items to be taken up at next meeting on Aug 24th.

Agenda

07/27/2016 Meeting #3 ( Cancelled)

1. Welcome to MANO WG formation team (5 min)

2. Agenda bashing and last meetings summary (5 min)

3.  Review the stated goals on Wiki - https://wiki.opnfv.org/display/mano (25 min)

4.  Inputs for preparation to present at subsequent TSC for providing/seeking any approvals and or guidance (15 min)

  • How to provide cross-project collaboration?  MODEL/VES/ECOMP are looking to work with Open Baton, Domino Looking to work with Use cases in OPEN-O, Edge Cloud Use cases KVM4NFV Low Latency/Multi-site  need guidance
  • Actual work (use case, gap analysis, interface requirement etc) needs to be driven and done within individual projects.

  • How to improve projects participation (OPNFV and Upstream) - Upstream has show great intrest like OPEN-O, OpenBaton

  • How to get  engagements from other projects ( Arm and Intel have both see play in VIM part of MANO Stack, we have from Arm Bob.Moankam@arm.com and possibly uri.elzur@intel.com  to join in this discussions)

5. How do we go about upcoming projects relating to MANO (5 min)

SUMMARY:

Meeting canceled due to logistics error.

07/14/2016 Meeting #2

1. Agenda Bashing

2. Discuss broad Goals -15 min

3. Review web page and see How Model Project can drive the MANO -15 min (This we could not as Bryan was not there and he discussed this with Guiseppe and have updated his input on Wiki)

4. How we engage with projects moving up the stack - Open Baton API - A point of view -15 min (API was presented and question on Endpoints, Generic VNFM vs. Vendor Specific, and ETSI standards and lack of some details like TOSCA Array for VL,VDU etc were discussed)

5. Action items and next meeting? (Doodle poll was suggested to get a fixed Semi monthly slot and improve participation)

Summary:

1.Some adjustments to 3 as Brian was not on call for Models, but he took up with Giuseppe and sent his input which is updated OPNFV Upstream Projects. 

2.Goals needed more Projects to participate internal as well upstream.before the message is sharpened.

3. First cut of Wiki was presented and Mano wiki needed changes (on-going - Mike, Bryan and Prakash to handle that)

4. As first presentation Open Baton proposed API based  testing of Interfaces and there are some variations in TOSCA NSD, VNFD, VDUl , VL etc and therefore API based testing needs Gap analysis.

5. Doodle Poll set for MANO Work Group on request basis.

Summary:

Although meeting minutes show only two projects Domino and Movie were interested more like MODEL, OPNFV (OPEN-O), Upstream Open Baton, JuJu and Tacker showed support and thus we proceeded with seeking participation.

06/30/2016

TC Minutes 20160630

06/21/2016 Meeting #1

Berlin Summit (MANO Discussions)

First discussion on MANO WorkGroup in weekly Tech meeting was based on  meeting with OPEN-O, Tacker & JOID teams and other community members and individuals interested in MANO_WG at OPNFV Berlin Summit and other OPNFV community meetings,  suggest the following as Scope for MANO_WG.  

Suggested Scope of MANO_WG in OPNFV discussed at

High level goals are

 A> Try avoid Divergence of evolving De-facto MANO WG for MANO Stack in OPNFV requirements  and ETSI MANO ISG General and evolving Normative specs.

 B> Help MANO evolution and innovation without being held hostage to slow evolving standards for OPNFV to lead the MANO integration.

 1. Scope of MANO_WG

 a) Identify from ETSI MANO ISG Specs few use cases that are most relevant to OPNFV service providers.

 (Only Service Providers participants to vote on  this and provide the IM/DM from Standards they prefer to use)

b) Core Interfaces to consider in MANO_WG to be based on API(and/or RPC messages) for identified use cases in a).

 [NFVO & VNFM Vendors to vote on this based on IM/DM provided in a) by SPs]

                I. VIM (API Produced by  VIM consumed by  NFVO & VNFM))

                II. VNFM ( API Produced by  VNMFM consumed by  NFVO & VIM )

                III. NFVO (API Consumed by NFVO from VNFM & VNFM based on use cases to be identified) 

3.  Review options for MANO system Information (IM) and Data Model(DM)  for supporting  Generic VNFM & Generic NFVO.

                - Define Generic-VNFM & Generic-NFVO based on required API/RPC and Interfaces to support standard Publicly used VNFs

                - VNFs format for provisioning  as required by VIM (Disk Image Builder/ Glance supported) and use cases agreed using  Generic VNFM

                - Any VNF Event, Notification  and Actions needed to support any Agents for VNF service required form use cases in a)

                - This will be agreed amongst all participating Vendors and SPs for MANO_WG in OPNFV

3. .  Alignment on MANO Application Information (IM) and Data Model(DM)  for supporting  accepted use cases using Generic VNFM & Generic NFVO.

                - The use case flow will be commonly accepted for all NFVO & VNFM based on ETSI ISG MANO General specs

                - Will also be verified by Test group projects to be able to test and support these use cases

                - A common reporting format will be proposed and used for all NFVO & VNFM working with OPNFV FuncTest & Yardstick  to pass testing requirements. This can also be used for Interoperability testing in Plug-fests.

4.  No item in MANO WG will be binding on any Vendor. The idea should be to facilitate options that can be practical and implementable yet allow innovation in emerging MANO stacks for OPNFV. Also anything that help Vendors to leave generic work in OPNFV the better and they can make specific variations in their Product lines that are super-set of the generic versions. 

5. This arrived MANO WG de-facto standard should be applicable in Release-D at the earliest.

Based on D Release experience OpenBaton team like to start integration from beginning of E-Release
OpenBaton will target  Doctor for E version, Needs documents ( Colorado is here http://artifacts.opnfv.org/doctor/review/23963/releasenotes/index.html)
For D-release please check with PTL Ryta Mibu ( https://wiki.opnfv.org/display/meetings/Doctor )

 

 

Aug 16, Wednesday : 14.00 UTC /7.00 PDT  Meeting #41 and summary

1.Agenda Bashing

2. F-Release OPNFV   based on ONAP related proposals in OPNFV

Discussion on Auto and Opera projects new proposal in pipeline and decision to re-scope and add new committers.

refer escalation Agreed in Tech meeting on Thursday - Action Tina Committers update for AUTO?, Action Yingjun Li on OPERA proposal updates?

3. E-Release  Progress in OPNFV  for Orchestra - Status and any conflict resolution sought?

(MS5 Scenario,MS6 Testing) based Orchestra for OpenBaton - - Any inputs on openIMS VNF testing from Orhcestra, JOID to OPNFV in Release-E

   a. Orchestra-15 (Michael Paul) ( Two step Builds for MANO - How?) -  Orchestra16 - Closed

   b. Functest Integration for vIMS testing / Reviewing MANO Integration tests from OpenBaton for OPNFV. - Status?

4. Discussions on VNF onboarding using VNFD & VNF Packaging best practices used by mano stacks (refer upstream)

    - Need comments from Auto/Opera/Installers and Orchestra projects in OPNFV (Action- JIRA ticket in Auto/Opera/Orchestra?)

5. Any updates on Best-practices  for adding Dynamic discovery of Mano components (Bryan/Uli) - No inputs to date should we close this?

Summary

Aug 9, Wednesday : 14.00 UTC /7.00 PDT  Postponed due to Bridge conflict.

Aug 2, Wednesday : 14.00 UTC /7.00 PDT  Meeting #40 and summary

1.Agenda Bashing

2, Release E updates for MS5 and MS6 from mano participating teams ( PTLs)

   a. Orchestra-15 (Michael Paul) ( Two step Builds for MANO - How?)

   b. Functest Integration for VoLTE testing (Debyan/Narinder) / Reviewing MANO Integration tests from OpenBaton for OPNFV.

3. ONAP-Automated OPNFV (updates - Tina/Bryan) - Opera Scope - Helen/Yingjun Li (Invited)

4. Dovetail - Testing gaps in FUT(functions)  for SUT (NFVI+VIM) - Srikanth Vavilapalli - Ericsson (Invited)

5. Any updates on Best-practices  for adding Dynamic discovery of Mano components (Bryan/Uli) 

Summary

opnfv-mano-wg-report-july-2017

July 26th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #39 and summary

1.Agenda Bashing

2. Best Practices - Options to consider MANO orchestration  in E-Release

      a.Two step MANO Orchestration wiki page  (Narinder)

      b. one step work-around for mano-descriptor until  then (Uli)

3. Orchestra OpenIMScore Charm review along with vIMS (Debyna)

4. Release E updates for MS5 from mano participating teams (all PTLs)

   a. Orchestra-15 (Michael Paul)

5. ONAP-Automated OPNFV (updates - Tina/Bryan)

summary

July 19th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #38 and summary

1.Agenda Bashing

2. Test - How to define common test for MANO based on scenario-complex-ncat for general dummy nfvo-vnfm testing in OPNFV

3. Opera Team changes to scope any and need to review? and seek approvals? / ONAP-Automated OPNFV

4. Scenario Descriptor updates on MANO for example1 based on Orchestra..(Narinder/Uli)

5. Best Practices - What to consider in E-Release (for testing based on 2 & 4 above)

Summary

July 12th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #37 and summary

1.Agenda Bashing

2. Orchestra Release plans - JOID/PoDs/Builds/Scenario/Deploy/Test -Orchestra/JOID Teams,

3. Opera revival & ONAP multiVIM requirements  / Laas -Opera Team changes to scope any and need to review? and seek approvals?

4. Scenario Descriptor on MANO use case to choose for it / Sample VNF team inputs.

5. Best Practices - On-boarding , Packaging, LCM (Orchestra vs Opera base on ONAP)

6. Use case discussions from Summit...VoIP Presentation

summary

July 5th, Wednesday : Meeting Canceled due to July4th Holiday week in USA.

June 28th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #36 and summary

1.Agenda Bashing

2. Revise mano Wiki pages for Release D+E(+F) targets + plans, besides  VNF Packaging and On-boarding by participating projects in different releases.(ALL PTLs)

3. Create and Deliver Documents from MANO Projects for providing their approach to MANO - Agree on common items to present or a template addressing EUAG Pain points and VNF Reference presented in Weekly Tech Meeting (.All PTLs)

4. Discuss and provide inputs for .MANO Scenario Descriptor (Ulrich Kleber)

5. Use case discussions for E /F releases for mano Integration.

summary

June 21st, Wednesday : 14.00 UTC /7.00 PDT  Meeting #35 and summary

1.Agenda Bashing

2. Review Wiki updates for Release D+E(+F) completions + plans, besides  VNF Packaging and On-boarding by participating projects in different releases.(ALL PTLs)

3. China Summit summary  for  MANO WG to share -  Multi-Access-Edge, TBD - https://www.opnfv.org/opnfv-summit-2017-event-recap

4. Create and Deliver Documents from MANO Projects for providing their approach to MANO - Agree on common items to present or a template addressing EUAG Pain points and VNF Reference presented in Weekly Tech Meeting (.All PTLs)

5. Discuss and provide inputs for .MANO Scenario Descriptor (Ulrich Kleber)

summary

June 14th, Wednesday : Meeting Canceled due to OPNFV China Summit

June 7th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #34 and summary

1.Agenda Bashing

2. Wiki updates for Release D+E(+F) completions + plans, besides  VNF Packaging and On-boarding by participating projects in different releases.(ALL PTLs)

3. China Summit preparation by  MANO projects and any presentation/ feedback for/from speakers.- MANO WG Project PTLs Invited  to present anything they plan  for Main Summit  (Bryan Sullivan) /  

Beijing Design (Prakash Ramchandran)

4. Create and Deliver Documents from MANO Projects for providing their approach to MANO - Agree on common items to present or a template addressing EUAG Pain points and VNF Reference presented in Weekly Tech Meeting (.All PTLs)

5. Discuss and provide inputs for .MANO Scenario Descriptor (Ulrich Kleber)

summary

May 31st, Wednesday : 14.00 UTC /7.00 PDT  Meeting #33 and summary

1.Agenda Bashing

2. Containers Packaging and On-boarding considerations for nfvo and vnfm + common practices checklist to consider

3. MNAO WG - focus areas to  facilitate Delivering E release priority & plan for F

4. China Summit preparation by  MANO projects and any presentation/ feedback for speakers.

summary

May 24th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #32 and summary

1.Agenda Bashing (Action Item review from last week)

2. Release E  followups on MANO projects plans and inputs from PTLs

  a.ONAP/OPEN-O Harmonization plans /Opera Readiness :  E-Release ( Action Item Closed -See Updates from Opera Yingjun Li and  ONAP acting release manage Gildas on ONAP)

  b.Orchestra has committed for  Build Deploy for E-Release and two JIRA tickets are opened (updates from Giuseppe -Await for his availability)

  1. Orchestra-1
  2. Orchestra-2

  c. Write a Juju charm for installing Open Baton and ONAP / Test Juju charm - (Action Item closed)

 d. Are projects in MANO WG  addressing EUAG Pain Points? Sample VNF is taking on the VNF Null(L2Forwarding), Echo.and what will the VoLTE, vCPE  (Moved back to Tech meetings - Action  Item Closed)

 f. How do we get to work in OPNFV for projects on F-Release those that missed E-Release window mid-cyle ? looking from community inputs and suggestions?

  JOID plans to support OpenStack Ocata with noSDN scenario for OpenBaton in release-E. (Action Item Closed )

  If ODL support for charm is added later it may consider that for OpenBaton at that time and Giuseppe to take it further for OpenBaton/Orchestra))

  d. Review of other projects in MANO WG that have indicated intent to participate in E-Release include Domino, Parser, Models and Copper (refer to f above for closure)

  e. Impact of kubernetes orchestration and containers on MANO Integration (Impact/Status of Helm in E-Release - no plans till Aug 3 and Action Item moved to -  Infrastructure WG -Giuseppe / Uli to start discussion on container scenario with LXD in JOID.

  f. Will any projects use Pike features for OpenStack in MANO integration? -None heard in OPNFV to date.

3. Discussions on Open Source SBC use case from MANO for Design and Deployment - Moved to Models-Domino and irc only meetings on Monday's  14 UTC/6 AM PDT #opnfv-domino

4. Events and planning MANO issues

OPNFV(Beijing) - Action on Demo PoC VoLTE(Moved to JOID meeting)

OPNFV Meetup (WSO2) and APIs (ONOS/ONAP)

5. Continuing discussions on VNF On-boarding, Packaging  best practices for MANO rationalization over OPNFV and inputs on ICE.

summary


May 17th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #31 and summary

1.Agenda Bashing

2. Release E  followups on MANO projects plans and inputs from PTLs

  a.ONAP/OPEN-O Harmonization plans /Opera Readiness :  E-Release (Updates from Yingjun Li on what are the issues for deferrals)

  b.Orchestra has committed for  Build Deploy for E-Release and two JIRA tickets are opened (updates from Giuseppe)

  1. Orchestra-1
  2. Orchestra-2

  c. Write a Juju charm for installing Open Baton / Test Juju charm - (other inputs from Narinder)

  d. Are projects in MANO WG  addressing EUAG Pain Points? Discuss Compile by projects what each project is doing for the same for weekly Tech meetings.

  JOID plans to support OpenStack Ocata with noSDN scenario for OpenBaton in release-E to start with. If ODL support for charm is added later it may consider that for OpenBaton at that time.

  d. Review of other projects in MANO WG that have indicated intent to participate in E-Release include Domino, Parser, Models and Copper

  e. Impact of kubernetes orchestration and containers on MANO Integration (Impact/Status of Helm in E-Release)

  f. Will any projects use Pike features for OpenStack in MANO integration?

3. Discussions on Open Source SBC use case from MANO for Design and Deployment 

4. Events and planning MANO issues

Openstack (Boston) - Reports & OPNFV(Beijing) -plans and OPNFV Meetup (WSO2) and APIs (ONOS/ONAP)

5. Continuing discussions on VNF On-boarding, Packaging  best practices for MANO rationalization over OPNFV

summary

May 10th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #30 and summary

1.Agenda Bashing

2. Events and planning MANO issues

Openstack (Boston) - Reports & OPNFV(Beijing) -plans and OPNFV Meetup (WSO2) and APIs (ONOS/ONAP)

.summary

May 3rd, Wednesday : 14.00 UTC /7.00 PDT  Meeting #29 and summary

1.Agenda Bashing

2. Events and planning MANO issues

Openstack (Boston) & OPNFV(Beijing) and OPNFV Meetup (WSO2) and APIs (ONOS/ONAP)

Due to change in Goto meeting  (https://global.gotomeeting.com/join/229632485 ) the meeting had a break and only two topics were covered.

We will use this new Bridge from next week and is also listed here

Summary

April 26th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #28 and summary

1.Agenda Bashing

2. Release E  followups on MANO projects / PTLs

3. Updates for new LF API Tool - Chintana (WSO2)

4. Prepare inputs  for Dovetail test qualifications for selected use case  from Models. (Model team/Prakash/)

5. Should MANO WG wiki be updated for ETSI updates -Release 2 presented by (Pierre Lynch)

6. Events and planning MANO issues for Openstack (Boston) & OPNFV(Beijing)

Summary

April 19th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #27 and summary

1.Agenda Bashing

2. Release E  followups on MANO projects / PTLs

  • Project PTLs inputs to E-Release content (VNF Onboarding & Deployment focus as Bryan Suggested)
  • Code Review  - Opera/Canonical  team members
  •  TOSCA for MANO scenario descriptor file from last week discussions - Bryan/ Narinder/ Uli - Await inputs
  •  SBC Opensource license inputs from LF
  •  SBC with 3-node VES / Hello-Tacker and VDU BP layout

3. Updates for new LF API Tool - Chintana

4.  Should MANO WG wiki be updated for ETSI updates -Release 2 presented by (Pierre Lynch)

April 12th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #26 summary

1.Agenda Bashing

2. Code Review report of vIMS testing in D1.0 using FUNCTEST - Orchetsra (openbaton as Orchestrator instantiating and testing   openims) - Moderated by Prakash , Giuseppe & Arturt to validate. - Code Review postponed, Dovetail qualification criterion for vIMS VNF testing is immature to qualify for testing at this stage, Bryan suggests focus on VNF Onboarding & Production Deployment for E-Release refer VNF Onboarding

3. TOSCA for MANO scenario descriptor file from last week discussions - Bryan/ Narinder/ Uli - Await inputs -Postponed

4. mano - testing use cases for various vnf types - Arturt - (What will be new adds besides SBC?, vOLTE?, vEPC?) - can be closed

5. Considerations for lower complexity  SBC use case based (define the use case first) on 3-node vHello Tacker in Modles Project (Bryan/Aimee/Prakash) - First Code Review of 3Node v Hello Tacker - BP + Start - Bryan suggested look at VES diagram and may be define the narrow use case for SBC using that for one external, one tenant facing and edge proxy using SBC to reuse the VES model. Plus for SBC as long as we donot modify can we get OK form LF for using this, route through Ray Pike.(Prakash to follow up)

6. Anything new from ONS summit at Santa Clara (April 3-6) on ONAP? - Inputs from Attendees any? - First presentation of ONAP occured at ONS. There is a meet for select ONAP team in coming weeks and we will hear more on that later.

7. Any followup  from MANO WG teams like WSO2 API  - Chintana (Updates for new LF API Tool ?) Chinta reported that they can import any API including Swagger and will review API tools next week as follow up.

8. Events  Plugfest (Paris),  OpenStack Boston + OPNFV Beijing Summit - Any updates? - Plugfest access request must be sent to morganDOTrichommeATorangeDOTcom

9. Should MANO WG wiki be updated for ETSI updates -Release 2 presented by (Pierre Lynch) -deferred


 

April 5th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #25 Summary

1.Agenda Bashing

2. Q1 MANO WG report - Prakash

3. TOSCA for MANO scenario descriptor file from last week discussions - Bryan/ Narinder/ Uli

4. testing vIMS feedback from test WG - Bryan/Prakash on review in Test WG, Orchestra (Giuseppe), Yingjun Li(Opera),

5. Considerations for lower complexity  SBC use case based on 3-node vHello Tacker in Modles Project (Bryan/Aimee)

6. Anything new from ONS summit at Santa Clara (April 3-6) on ONAP?

7. Any followup  from MANO WG teams like WSO2 API and  for Events  Plugfest (Paris),  OpenStack Boston + OPNFV Beijing Summit

8. mano testing use cases for various vnf types - Arturt

Summary of meeting#25 

March 29nd, Wednesday : 14.00 UTC /7.00 PDT  Meeting #24 summary

1.Agenda Bashing

2. Discussions & Actions on Process aspects and API for VNF vIMS Build / Test in  FUNCTET and its value for Release  E. .

3.  Use case discussions  -  vOLTE, can we simplify this?

Summary of meeting

March 22nd, Wednesday : 14.00 UTC /7.00 PDT  Meeting #23 and  Summary

1.Agenda Bashing

2. Discussions & Actions on  MANO Build Options in Release-E - Giuseppe , Yingjun , Arturt, Ulas

3. Streamlining VNF Onboarding & LCM Release-E  and Action plans - Giuseppe , Yingjun , Narinder, Sameer

4. Time permitting Use case discussions -   from Models Test and Samplevnf, vOLTE (Opera request) - deferred for Open-O and Models co-ordination.(Pakash, Yingjun Li, Bryan)

5. Followups on items from last week.

Summary of meeting:

NFVO and VNFM both are external components and OPNFV build gets these files as software artifacts, bundles or as images (VM or containers)

In release D 1.0 the Opera chose to use an openo scenario and Compass Installer and Orchestra uses direct Orchestration from FUNCTEST.

We will have more details once D1.0 is released as what makes in D1.0 and what goes to D2.0

March 15th, Wednesday : 14.00 UTC /7.00 PDT  Meeting #22 and summary

1.Agenda Bashing

     Refer log of Tech meeting last week  MANO WG update - http://ircbot.wl.linuxfoundation.org/meetings/opnfv-meeting/2017/opnfv-meeting.2017-03-09-14.01.html

    Discussed E-Release schedules  and listed that Slides for E-Release planning in 5 below.

2. Review action items from last meeting and progress on  - deferred

     AI101/102/103 - Scenario, MANO Components , External vs. Internal Build  - PTL Opera/Orchestra

     AI201/202 - VNF On-boarding Flow + Use case descriptors (vEPC/vIMS) - PTL Models/Domino - Prakash

3. New timings for Models+Domino for inputs to MANO? - Informed

4. Review MANO WG plans for OpenStack Boston +  MANO WG Survey Questions? -Deferred

5. Discussions on work items identified for Q1 not able to cover from last 2 meetings  if time permits - Discussed as part of slides below

Slides for E-Release Planing

Summary

Delayed start by 10 minutes on GTM because of connection call start issues.

Yingjun Li - Opera  may need extension for Intent to participate in E deferred because of ONAP (OPEN-O & ECOMP harmonization)

Atrurt - MANO integration need 3 pieces: NFVI, MANO and VNF involvement plus sync with Functest

Debyang/Sameer -  Presented updates for vCPE/vIMS based on OAI and believe they can help test and Demo with Orchestra through participation with MANO projects

Giuspee - We need to clarify process flow of Build and deploy over Scenario(s)

Prakash - Agreed to start a thread based on Giuseppe request summarizing the discussions here in

March 8th, Wednesday : 15.00 UTC /7.00 PST  Meeting #21 and Summary

1.Agenda Bashing

2.MANO WG Survey (TBD) / MANO WG Feb report

3. Models/VNF Graph additions from Models/Domino call from this week for use cases through Models and possible SampleVNFs

4. New timings going back to 14.00 UTC ( 7 AM PDT, 9 PM EU, Asia India  IST 7.30 PM / China CST 11 PM)

5. Discussions on work items identified for Q1 not able to cover from last week.- 

Refer log of mano-wg meeting 

http://ircbot.wl.linuxfoundation.org/meetings/opnfv-meeting/2017/opnfv-meeting.2017-03-09-14.01.html

March 1st, Wednesday : 15.00 UTC /7.00 PST  Meeting #20 & Summary

  1. Agenda Bashing
  2.  JIRA for consolidating E-Release Priority : option 1 - Have separate one for MANO WG option 2- Use the ones by respective Projects to refer to - For now option 2 will suffice till we have clarity on ONAP(ECOMP+OPEN-O Harmonization to use Action Item AI)
  3. ONAP and its impact on MNAO WG for OPERA (OPEN-O/ECOMP) - ONAP slide 1.0
  4. Final Agreements for Release E by both Orchestra , Opera and other participating projects + Upstream

Agreements Approval by MANO WG team

Consolidated E-Release Work Items from MANO WG

MANO Projects in E release will focus on test tools for integration and testing MANO stack  from Meeting #18

 Agreed consistency across MANO stacks we mean "Consistency of end users experience" from Meeting #18

Work Items and action items to follow up based on Priorities: ( we will discuss this next week in details)

Priority 1 is Integrating MANO Stack Components over OPNFV VIM from Meeting #19 2017

AI-101 What Scenario MANO Project / General Scenario Reveiew will use in E?  Create  /  Generic or    Specific :Ideally would be use Generic or None. Read Document for CI - Section 7 MANO Scenarios.

Note Both top-down and bottom-up approach must meet  generic OPNFV VIM scenario and some inputs from EUAG for vCPE and NaaS use case for scaling is worth noting.

Should MANO Projects consider : Parent-Child Scenario?

Project::Scenario/Installer/AAA

CI evolution: Boxing Installers into VM on Jumpservers, Should we box NFVO and VNFM into VM on Jumpserver or Compute nodes?

Pharos Specification Update for MANO: Should MANO WG update Pharos Specification last update was update was for Colorado (PHAROS-144)

Copper: none

Domino: os-nosdn-nofeature-ha /JOID / keystone v2 / post-install

Models:

Opera:: os-openo-nofaeture-ha/ compass & JOID/ keystone v3.

Orchestra: not defined

Parser:

AI-102: How do you plan to add your NFVO/VNF or Feature components : Project: Components/Feature review parent/child with os-mano-vxlan-ha as parent and other like opera or orchestra deriving from it as child till difference are settled or we can drop scenario totally.

AI-103: Will you use upstream Build and just provide artifact on a path or URI or use Internal CI evolution to dynamically build and test over vPODs or PODs in OPNFV Community? Describe your MANO Project preference and flow chart for evaluating processes or refer your code.

Priority 2  is considerations for agreement on VNF On-boarding from Meeting #19 2017

AI-201: Describe what is your function flow for on-boarding  (be simple Hello world of vIMS VNF) - We have common ways on on-boarding with some variations  and a common flow or metadata to use?

AI-202: Do you use VNFD(with NSD) or just a simple Blueprint (BP) or TOSCA node template (review  vnf2 yaml file)

Priority 3 is Continuing on Reviewing Best Practices for VNF On-boarding & LCM from Meeting #19 2017 plus EUAG reference

AI-301:  Document or point to vIMS On-boarding procedure in upstream for a) OPEN-O b) OpenBaton c) ONAP/ECOMP

Priority 4 Use inputs from OSM for VNFD and NSD challenges. from Meeting #15 -2016

AI-401: Document and present at MANO WG what we need to l;earn from OSM?

Volunteers who can work on AI on Priority 1 & 2 first.

 

Feb 23: After last meeting of MANO on 22nd a new LF project called ONAP has emerged merging OPEN-O + ECOMP efforts and we cover it on March 1st meeting. See above.

Feb 22, Wednesday : 15.00 UTC /7.00 PST  Meeting #19 and summary

  1. Agenda Bashing
  2. Action Items for last meeting and closures for this month for E-Release Content
    1. Action Item 1 (AI1) of last meeting see Wiki Page for MANO-Best Practices  added besides and needs priority for E agreed here by all Participating projects.

      (Giuseppe - OpenBaton) offers Packages for VNFM for gVNFM as well Juju. Docker based packaging is also being considered beyond OpenBaton 3.0.

      Bootstrap for NFVO+VNFM will be from OpenBaton and Integration with OPNFV VIM will be using JOID Installer + Scenario (to be introduced in E).

       (Yingjun - Opera)  Scenario for OPEN-O os-openo-nofaeture-ha with compass & JOID with keystone v3.

      AGREED: Priority 1 is Integrating MANO Stack Components over OPNFV VIM (Test area 1 will address this in Dovetail through  tests to be proposed for MANO and also the MANO Component Configuration  to expand scope over OPNFV VIM Scenarios.)

    2. Considerations for agreement on VNF On-boarding (Priority 2)

      Refer to VNF OnBoarding and within that End User Advisory linkReview item 4,5,6 from last meeting and Interns 5 priority for VNF Catalog  
      https://lists.opnfv.org/pipermail/opnfv-tech-discuss/2017-February/014950.html - Bryan
      + Based on email reply Prakash
      VNF should establish its name and context as an Application or Component with a path to Application/Component Catalog(Cat.3/Repository). This assumes that hierarchy of Catalogues/Repos is Cat.5/Repo5 (Marketplace), Cat.4/Repo4 -Service Provider Service Catalog (This is where you on-board NS/NSD), Cat.3/Repo3 - Application/Component Catalog (This is where you on-board your VNF/VNFD/Packages etc), (Cat2./Repo2) -SDN Controller Catalog. Cat1/repo1 is VIM/Glace Catalog. Catalog Cat0/repo0 will be instance Catalog. e.g.. HEAT.    
      (Giuseppe) Define first the metadata and content like using Blueprint etc before a formrat can be made or agreed.
      How you create package is through SDK which is being proposed by both OpenBaton and Open-O 
    3. Plugtest and inputs for MANO from Participants to see if anything can be used for E-Release content - Since there was a nondisclosure agreements by participants this item is closed for now.

  3. Code review for VNF testing - Completed  common testing for vIMS (Orchestra to use Orchestration and Opera to use Compass Installer for testing vIMS as VNF use case). Added request for Onbord enabled flag and externalization and add to Service Catalog of enabled, Package name/type and descriptor explicitly through extended class to baseVNF class as per new files for Opera and Orchestra.
    What will be selection of use case and test areas/tests which we can drive MANO inputs for D release of Dovetail (vIMS - Test Area 2 - suggest adding vEPC to existing vIMS and improvise to add End to End Packet Data Besides Voice services- needs input from Opera and Orchestra.

    Continuing on Reviewing Best Practices for VNF On-boarding & LCM (Priority 3)
    a. MANO SB APIs Or-Vi is most mature at this time to consider for E-Release
    b. Close loop monitoring is a possible candidate for OPEN-O with Doctor Project (Giuseppe for OpenBaton in E-Release)
    c. Go beyond to adding Policy for VNFD as use of Simple TOSCA YAML file which is current norm for most MANO projects. 

  4. OSM & ETSI NFV updates - Refer [ RP ] in Meeting #15 -2016.

Feb 15, Wednesday : 15.00 UTC /7.00 PST  Meeting #18 continued from last week and summary

  1. Agenda Bashing
     
  2.  Report from Upstream  integration - OPEN-O /Opera - Yingjun Li
  3. Report form OPNFV participating project PTLs  for better Orchestration & Automation i E-Release.
    1. Models Meeting#24:  - Models to focus on completing Blueprints of VNFDs support in E-Release. Visibility to Upstream limited, high level abstractions like in OPEN-O not visible to OPNFV - Info Models support from ARIA, OPNFV, OpenStack, Kubernetes etc. At test level will continue  to shift from Stable to  Trunk - Continuous Development on upstream Trunks. Primarily OpenStack and beyond like OPen-o, ONOS, ODL, OpenContrail etc.
    2. CopperGot Congress support in Installers but  ODL GBP lack of contributions.  Continue to support E-Relase but may Sunset in D time frame, and move everything upstream, Remaining FUNCTEST etc be in upstream, deployment should possibly be  installer agnostic.
    3. Parser: Await inputs from OPNFV Parser, Inputs from ARIA.
    4. Domino: Evaluating Integration for NFVO Registration with OEPEN-O/GSO, integration with OPEN-O GSO using TOSCA and policy , considering multi-parser etc.
    5. JOID(LXD): Workloads of K8 for LXD in E. link https://bugs.launchpad.net/charms/+source/percona-cluster/+bug/1663245
    6. VES :  Analytics in closed loop policies , Barometer has few and therefore Integrate and test using VEs & Barometer. Another possibility is Redhat ManagedIQ. Use Cases like Fault, scale, push, pull, performance requirements
  4. Review and Guidance for MANO based on newly specified OPNFV Architecture - OPNFV Architecture for Danube release - New from TSC call 1/31/2016

Feb 8, Wednesday : 15.00 UTC /7.00 PST  Meeting #18 and Summary

  1.  Agenda bashing
  2. Follow-up on Action Items and  from last month and finalizing E-Release Goals and agreements
    1. Goals for E should be to release test tools for integration and testing MANO stack -Bryan
       

      E-Release Goals and Contents

       

      E release must focus on test tools for integration and testing MANO stack

       

      Agreed that when we mean Consistency across MANO stacks we mean "Consistency of end users experience"

       

      This means each MANO stack may follow different formats and methods but process as it appears to an end user is consistent.

       

      AI 1: Compile VNF Formats , Onboarding process, Catalogs and LCM for few stacks to understand the differences and try align for consistent end user experience.

      (Prakash will initiate a Wiki page and Bryan and others will edit and update it for next meeting to go over rest by emails)
    2. Release E discussions: using keystone Auth v3 in Release E - Moved from D to E -D2E - Guidance is OK and its no binding and will depend on the progress of each MANO Stack and its interaction with installer.and OpenStack.
    3. Review and Guidance for MANO based on newly specified OPNFV Architecture - OPNFV Architecture for Danube release - New from TSC call 1/31/2016 -This could not be taken up deferred for future meetings.
    4. Report from upstream (OpenBaton, Open-O, openECOMP, JuJu, Tacker, OSM/Plug-test) and its impact on MANO WG - Details of Inputs from MANO Projects refer below Details of Discussions in Meeting#18

  3. How to arrive at best practices for NS and Cloud Native VNF Package Formats and run time, - Deferred to next meeting- Add to AI 1 for next meeting to review on Wiki
    1.  Generating YAML(TOSCA/YANG)  refer link  Models VNF-Packaging and Riftt.io , TOSCA file inside CSAR zip file
    2.  Linux Package-Management formats , OpenSUSE Package Manager
  4. Follow-up on Action Items and  from last month and finalizing E-Release Goals and agreements (must be before March 27 as per EUAG) - Add to AI 1 for next meeting to review on Wiki
    1. NS/VNF Package Standards adapted ( TOSCA/ CSAR (OPen-O)?, TOSCA / rift.io Juju YAML (OpenBaon?), TOSCA/Rift.io (OSM), ECOMP ? - AI who?
    2. VNF On-boarding Process across MANO stacks (to be consistent? how?) - AI (who?)
    3. Keystone v3 - (Installers JOID/Junju in E?)
    4. Use case continue with simple Hello world + vIMS or add one more in E?
  5.  Plugtest and inputs for MANO from Participants. - Deferred to next meeting

    Details of Discussions in Meeting#18
    OpenBaton : Guiseppe on Architecture comparison
    We have one done as it exists and evaluated in January 2017: https://wiki.opnfv.org/display/mano/Architecture
    OpenBaton has  Project dependency - Need help  from JOID and Mile Stone support (MS?). Feature projects don't have all Milestones to meet
    David McBride is working in Release Management team to simplify the Milestone process to be separated for Specification, Feature Projects, Installers etc.

    Being a Feature Project and working through FUNCTEST in D Release the support is limited to FUNCTEST of vIMS use case
    JujU and JOID are not same, JujU is a GVNFM and Canonical does work with OpenBaton.
    Issue is more of Integration and testing with FUNCTEST whic uses Swagger APIs, and Morgan has been helping OpenBaton.
    The FUCTEST team has reduced pain point for OpenBaton by providing Orchestration option in FUNCTEST for VNF (vIMS)
    Similarly FUNCTEST has allowed Open-O to use Installer option through JOID and Compass and for Compass Open-O is using Keystone V3 in its scenario.
    Based on D Release experience OpenBaton team like to start integration from beginning of E-Release
    OpenBaton will target  Doctor for E version, Needs documents ( Colorado is here http://artifacts.opnfv.org/doctor/review/23963/releasenotes/index.html)
    For D-release please check with PTL Ryta Mibu ( https://wiki.opnfv.org/display/meetings/Doctor )

    JOID/JuJu Canonica:l Artur/Narinder mentioned that Charm for Orchestra is ready and should be able work throug JOID too and OpenBaton can work with JOID team.
    Testing code for vIMS is getting ready, link https://gerrit.opnfv.org/gerrit/#/c/26769/
    Artur from Canonical like to see in Release E, use of JuJu and Charms in  Open-O, OSM, OpenBaton part of Functest  Scenarios etc.

    Cross Project Dependency
    Can we invite Doctor PTL or Committer to present in MANO WG as how to integrate with different MANO stacks? -follow up through email for now.

    For Models what can be used  Giuseppe and Bryan will work out using Python once OpenBaton has some codes to use in Gerrit in Orchestra project.

    Discussion about scenarios and features...
    architecture - management models NB
    Potentially there will be test that will integrate into ECOPM , are working on VES
    How MANO projects address how open-loop and close-loop control and telemetry , LCM , using Celometer/Heat, Redhat has Insight and Satellite 6  , the other is Barometer in OPNFV etc
    Add to AI1 above : Service creation., onboarding, overall orchestration, Application LCM, Inventory system, Analytic (Close loop system), Plugin like Policy, Security and Resource Orchestration etc.
    Action Item for Prkash to create a wiki table..Prakash and Bryan will support it to enlist items in AI 1 here and above.

    Sridhar K.N. Rao /Solution Architect wanted to know if test tools in Release E refer to Testing Projects OR open Source tets tools like REST API for VNFM , testing Framework like Dredd?
    https://dredd.readthedocs.io/en/latest/ & https://apiblueprint.org/tools.html - May be ask for a presentation from Redhat or Sridhar to present this if MANO WG consider this as a value for E-Release
    Bryan Sullivan answered : The point being made in that note was the we should be working tactically to setup test environments for MANO stack testing now and thru the E release.
    I do not believe that in the next 6 months we will have the ability to drive convergence in any real sense, except thru use case testing with existing open source MANO solutions.
    In the Models project we are working to establish a pattern for such tests, focused on assessing the set of blueprint features supported by various MANO stack components.
    There is a similar effort underway as mentioned by Giuseppe in the Functest project.
    I had recommended earlier that test approaches to MANO functionality should be in the scope of the MANO WG, and that Functest should simply incorporate what the MANO WG develops as a framework.

    But in the meantime progress follows momentum, and that momentum is in the Models and Functest projects.

    Meanwhile we did not get inputs from Yingjun Li from OPEN-O /Domino/ Parser and like to schedule inputs at  meeting#19 before we conclude on E-Release updates
    Time lines: By Feb 23 to bring it to Tech Meeting before getting to TSC by Feb 28

Jan  25, Wednesday : 15.00 UTC /7.00 PST  Meeting 17 and Sumary

  1.  Agenda bashing
  2. Review of last meeting (#15) and actions
    1. Priority for D - completed and accepted
    2. Review of VNF on-boarding - Is anyone Interested in monitoring Intern for VNF On-boarding - No inputs yet
    3. Code review for VNF testing - Completed  common testing for vIMS (Orchestra to use Orchestration and Opera to use Compass Installer for testing vIMS as VNF use case) - To review feedback next meeting along with plugtest.
    4. Scenario conclusion - Orchestra , Opera, Domino , Model , Parser, - Do we have a path to no scenario? Need to fix a simple framework and refer  Models -see vHello_Tacker.sh and Danube Scenarios.
    5. Priority Release E (Move left out D to E + for consistency)- Goals for E should be to release test tools for integration ans testing MANO stack
    6. ABOT framewrok for BDD has impact beyond MANO and hence moved the discussions to Tech Meeting - Completed Transfer to Weekly Tech for Thursday.
  3. Architecture comparison and LCM.
  4. OSM & ETSI NFV updates - Refer [ RP ] in Meeting 15 -2016 -Deferred to next meeting after Plug-test
  5. Report from upstream (OpenBaton, Open-O, openECOMP, JuJu, Tacker, OSM/Plug-test) and its impact on MANO WG -Deferred to next meeting
  6. MANO related tests for Dovetail beyond C for D based on what we have and who will volunteer to get this moving? - Dovetail uses Functest but we should invite them to present their tool for benefit of MANO projects or to Tech meeting so we understand what they did in C to see how it will move up the stack in D and help MANO testing in future.
  7. Continuing on Reviewing Best Practices for VNF On-boarding & LCM based on current community project status? - Strategy towards tool deriving from Models like Cloudify , Tacker and will need more discussions with participating MANO projects to help in this efforts.
  8.  Actionable Items and Agreement summary recording - A1. Bryan for wiki Architecture page move, A2. Prakash for Release E update & Getting Dovetail Test tool presentation in MANO or Weekly Tech meeting.

Jan 11, Wednesday : 15.00 UTC /7.00 PST  Meeting 16 & Summary

  1. Agenda bashing
  2. Review of last Quarter (2016) work - Bryan Sullivan on what we did and what we missed (Set Priority) - Proposed VNF On-boarding and Pilots for testing and satisfied with progress
  3. Plan priority for D & E releases of OPNFV ( Review request for VNF OnBoarding for vIMS) (Code-Review or testing Vnf) - Agreed on D-Release Priority write up reviewed link and plan for E future meetings, Code Review lest to participating projects to comment on abstractions and ability to adapt by OPEN-O, OpenBaton and other teams. Some comments from Serge Manning (Sprint) on Inter-operability and Bryan indicated that goal is to move towards consistency and portability for providers & end users for VNF LCM etc. The AAA and Keystone V3 was discussed and David McBride Release manager pointed to link for same.
  4.  Common Test for vIMS (clearwater) - This was used as starting point for testing complex VNF like vIMS and code is being adapted for testing as in 3.
  5. Opera & OpenBaton Review (see links below) - Refer [ YC ] and [ GC ]  in Meeting 15 -2016 - Both Giuseppe and Yingjun Li represented Orchestra and Opera respectively and indicated continued participation in testing vnfm for D release. YC to allocate a code reviewer for testing vnf link form OPEN-O team, Giuseppe has reviewed and looks OK but needs more details to be included from the points listed in Meeting 15 content below.
  6. JuJu VNFM and mano Scenario support for JOID - Refer [AT ] in Meeting 15 -2016 - Artur and Narinder indicated OAI as EuruoCom initiative and also pointed to Deb from Rebaca.
  7. ABOT Framework feedback - Refer [ DC ] in Meeting 15 -2016 - Due to lack of time and also nature of DBB and testing it is considered this be presented at Tech meeting and we will check with Bin Hu if he can help get this session instead of MANO as it covers wider involvement of community to assess this for OPNFV.
  8. Architecture comparison and LCM - Deferred to future meetings.
  9. OSM & ETSI NFV updates - Refer [ RP ] in Meeting 15 -2016 - Deferred to future meetings.
  10. Any action plan for Q1 2017.- Just send the link of Release - D priority to community and see if this will suffice for Q1/2017.

Nov  30: Dec 7,14  2016 (Q4) Meeting 15 (Hackfest + Ad Hoc):

[ YL]  Yingjun Li (Opera) provided details of Hackfestprep and delivery

https://wiki.opnfv.org/display/meetings/Opera

https://wiki.opnfv.org/display/PROJ/OPNFV+OPEN-O+Integration

https://wiki.opnfv.org/display/PROJ/OPERA+Function+Test

[ BS ] Bryan Sullivan's Input from Hackfest for use of Secnario and Installations: refer inside to POD/Genesis support MANO vPODs

https://etherpad.opnfv.org/p/infra_dec_hackfest

[ GC ] Giuseppe Carella (Orchestra) provided details for OPNFV MANO Integration.

In general, in Orchestra we would be interested in having some integration scenarios which could be useful for covering test cases for MANO frameworks. We have already an upstream project [1] doing something similarto VNF Onboarding functest, and our initial plan was to integrate this project within OPNFV functest, as already mentioned earlier to Jose. With the approach taken in your vnf onboarding abstraction I see that this integration won’t be really easily doable. Main reason is that our integration-test framework already splits the test cases in different steps, while you cover at the moment mainly two steps (deploy VNF and test VNF)

* registration of a VIM instance 

* on boarding a NSD

* deployment a NSR using the on boarded NSD from previous step

* testing that the NS is really deployed (optional in case of negative testing)

* deletion of the NSR

* deletion of the NSD

* unregistration of the VIM

 

With our framework, it is possible to define a scenario (and the steps which needs to be executed) directly using a configuration file. An example of a scenario can be found here [2], deploying an iperf server and client network service and testing that the client connected effectively with the server (therefore testing that dependencies were correctly configured between VNFs). 

 

Although I see that your approach can be further extended to include those additional steps, I would suggest not to have them as part of VNF testing, rather starting creating a new set of test cases which can be part of a MANO group (or maybe including them in the feature tests group?). Nevertheless, coding the test steps in a specific programming language will also limit the integration of existing external frameworks (like the one we already have) MANO providers may already have implemented. 

 

In summary, talking about the Orchestra project, we will provide soon the code for testing the deployment of a VNF (vIMS based on OpenIMSCore) implementing the abstract class you provided and following the approach taken for cloudify. However, as we assume that in the future the Orchestra project will be directly integrated with at least one installer, we would like to include some more comprehensive use cases that can execute several (positive and negative) test cases, possibly using already existing frameworks (like the one I mentioned). Would that be feasible within functest? 

[ 1 ] https://github.com/openbaton/integration-tests

[ 2 ] https://github.com/openbaton/integration-tests/blob/master/src/main/resources/integration-test-scenarios/scenario-real-iperf.ini

[AT ]Canonical's and OAI input for VNFM

Canonical and OAI inputs to OPNFV MANO for VNFM

[ DC ] Debyang Chaudhari of Rebaca Automation on ABOT BDD test framework

ABOT BDD Framework

[ RP ] Ramchandran Parkash on following on VNFD and NSD information Model Challenges in ETSI NFV

VNFD Challenges

·         VNFD Connection Points

·         VNFD Deployment Flavour & associated changes in VDU topology

·         Lifecycle management events and scripts referenced in the VNFD

·         Enhanced Platform Awareness Attributes

NSD Challenges

·         Deployment flavour complexity

·         Nested service ambiguities

·         Local Affinity Rules vs Affinity Groups

·         VNFFGD and the lack of relationship with deployment flavours

·         Lifecycle management events and scripts references in the NSD


Changed to weekly meeting due to MANO activities in Q4 2016.

Agenda of meeting November 30, 2016  15.00 UTC /7.00 PST Meeting #14 and Summary

  1. Review of last week work and Agenda bashing

      2. The  FUNCTEST-353 is more comprehensive than  focus on API testing through common wrappers. Therefore  4 mores tasks are created under Functest to address that and they are

        a.Understand the architecture ( Inbuilt as part of OPNFV, Install as part of Orchestration the NFVO, Install and and test ).

        b. Adding NFVO to installer

        c. How to test or-vi - FUNCTEST-614,

        d. How to test or-vnfm -   FUNCTEST-615

       Yingjun/Giuseppe please ask your teams to  compare with https://git.opnfv.org/functest/tree/testcases/vnf/vIMS?h=stable/colorado this code with Java codes presented last week and give feedback.

     3. Action Artur to plan for getting Rebaca Automation team to present on how they use Juju with NFVOs - ABOT Presented and reviews still in progress

     4. How would the results of collaboration through MANO WG reflect in D-release of OPNFV. - A document template  to be proposed and  reviewed to see what should the best practices to include in the D-release based on MANO WG participation by Opera, Orchestra and associated VNFM.

     5. Discussions on what would be inputs from MANO_WG to OPNFV D & E Release. Wiki Page to update after discussions in WG. - Bryan to update this for Release D with minimum and Prakash to review all the participating projects and provide a report this meeting on 30th Nov.

     6. Any  Updates on UNH Plug-fest

Agenda of meeting November 23, 2016  15.00 UTC /7.00 PST Meeting #13 and summary

  1. Review of last week work and Agenda bashing

    Need FUNCTEST-237 replaced by one that Opera/Orchestra will commonly use - YingjunLi/Giuseppe)

    Giuseppe went over the codes listed for wrapper calls for or-vi and or-vnfm and explained how it uses Java Public calls to invoke VIMDriver which can be a Python Client or API calls to OpenStack VIM.

    https://github.com/openbaton/openbaton-libs/blob/master/vim-int/src/main/java/org/openbaton/nfvo/vim_interfaces/vim/Vim.java

    The code below is similar code for or-vnfm for VNF lifecycle management calling the rabbitMQ and can be aligned for calls to JuJu or tacker with some common agreements to be discussed next week.

    baton/openbaton-libs/blob/master/vnfm-sdk/src/main/java/org/openbaton/common/vnfm_sdk/interfaces/VNFLifecycleManagement.java

    For each of the  projects OpenBaton/Orchestra, Open-O/Opera please detail the items A,B,C and D below (Bryan to reword and fix the same below)
    A. Compile workflow of LCM  of a VNF.

    B. Compile  the  MANO  NFVO components/Blocks  for OPNFV MANO stack Integration. - Action Item on Prakash

    C. Define and describe the  role of VNFM (Juju, Tacker) if these are used in B as options. - Action Artur to plan for getting Rebeca team to present on how they use Juju with NFVOs)

    D. Compile Sequence Diagram for LCM of VNF through the components/Blocks (NFVO, VIMDriver, VNFM,...) - Giuseppe suggested this to be able to co-relate with codes links listed above.

  2. Where is the test plan and how will the projects share the testing codes? - Proposal to be discussed next week with inputs for Giuseppe of Orchestra (OpenBaton team). -Deferred to item 2/Nov30
  3. How would the results of collaboration through MANO WG reflect in D-release of OPNFV. - A template to be reviewed to see what should the best practices to include in the D-release based on MANO WG participation by Opera, Orchestra and associated VNFM. -- Deferred to item 3/Nov30
  4. Discussions on what would be inputs from MANO_WG to OPNFV D & E Release. Wiki Page to update after discussions in WG. - Bryan to update this for Release D with minimum and Prakash to review all the participating projects and provide a report next meeting on 30th Nov. 
  5. UNH Plug-fest for cross-project co-ordination? - Update from OPEN-O/Opera team - This Bryan is working with JuJu team to see if any VNF cam be demonstrated at UNH. Plus to avoid any sign-up issues the Plug-fest and Hack-fest discussions on MNAO are separated out.

Agenda of meeting November 16, 2016  14.00 UTC /7.00 PDT Meeting #12 and Summary

  1. Summary of last meeting and Agenda bashing - No additions
  2. Compare Clearwater vIMS vs. OpenIMS- (Artur Tyloc - details on OpenIMS and JuJu orchestration for the same)
    Bryan asked what comparison means and was agreed that we work to find on-boarding package of vIMS and workflow for LCM of vIMS.
  3. Review  or propose TOSCA template for vIMS and OpenIMS for On-Boarding and compare with ( NFV (VNFD, VLD, VNNFGD, PNFD) - (Yingjun Li) - the  https://github.com/Orange-OpenSource/opnfv-cloudify-clearwater (openstack-blueprint.yaml) has details. His team is working to understand and use it for now for UNH Plugfest.
  4. What VNF  package should  contain besides TOSCA descriptor ( for CW vIMS / OpenIMS) - OPEN-O & OpenBaton for their NFVO & VNFM choice they make (Opera/Orchestra teams) - snap from Canonical http://snapcraft.io/ was discussed and other one we had looked earlier was CSAR
  5. Service Invocation and Termination API versioning and flows for VNF (vIMS/OpenIMS) use case.(General Discussions) - Agreed that we must include verisioning while using API for LCM of VNF.
  6. Inputs from OpenBaton for IFA5,6,7 Interface functional calls. (Giuseppe) - Lorenz pointed to various nfvo/vnfm scripts in https://github.com/openbaton
  7. Progress on Models  based on Peter Lees inputs using Django & YANG. How will that apply to vIMS/OpenMANO? (Bryan)
    Bryan added link  for comparison - Comparison, references , testing


Agenda of meeting November 9, 2016  14.00 UTC /7.00 PDT Meeting #11 and summary

  1. What should we document for MANO best practice for OPEN-O and OpenBaton integration through Opera and Orchestra.
    1. What use case to start with - - Should vIMS be the first use case?
      vIMS is agreed use case to document as first one for MANO integration
      Besides Clearwater vIMS Blueprint in OPNFV,  Arthur from Canonical suggested using OpenIMS and JuJu can handle both.
    2. VNF On-Boarding - TOSCA Node Template <similar to> NFV (VNFD, VLD, VNNFGD, PNFD) - Onboarding
      OpenBaton / Orchestra indicated that they like to see VNF Package to include VNFD along with image name and script to instantiate the VNF. So did Yingjun Li PTL Opera chimed and thus there is agreement on this from both projects. The issue of formats to use include CSAR & JSON and run times both Python and Java are being considered. Thus there will be attempt by both teams to define them along with VNFM's they plan to use like JujU and Tacker.

      The VNF package, and data model / representations for NSD and VNFD, and how/why those vary across MANO projects

      b1.       OpenBaton: JSON for VNFDNSD; TOSCA for NSD

      b2.       Open-O: …Needs input from Opera team

    3. Service Invocation and Deletion - TOSCA Service  template = NFV NSD, How do we handle SO & RO for this?
      This will follow NFVO flow to VNFM asking to allocate resource and if it does not that capability then NFVO will reroute the request to VIM. On success of Resource reservation (Quota) etc. the NFVO will ask VNFM to instantiate the service for the deployment completion as first target. Life cycle management will follow that based on agreed hooks to be defined as common between the Opera , Orchestra, Juju and Tacker teams. A proposal will be reviewed next week.
    4. API versioning - which API?
      The API versioning is of abstractions or wrappers that is being proposed by OpenBaton and will be reviewed by other MANO participants to agree on a common wrapper approach for IFA 5, 6,7 and OPNFV VIM and VNFM drivers from NFVO.
  2. At what point should upstream integration project  consider OPNFV Models topology design and templates to choose  and use. - This will be discussed at next Models call with Bryan Sullivan.
  3. What should be test strategy (refer  item 3 Nov 2 meeting) - Await FUNCTEST-237 response from Morgan and Action for MANO teams to create new FUNCTEST JIRA ticket based on vIMS use case.
  4. Where is the test plan and how will the projects share the testing codes? - Proposal to be discussed next week with inputs for Giuseppe of Orchestra (OpenBaton team).
  5. How would the results of collaboration through MANO WG reflect in D-release of OPNFV. - A template to be reviewed to see what should the best practices to include in the D-release based on MANO WG participation by Opera, Orchestra and associated VNFM.
  6. What is reasonable deliverable expected from UNH Plug-fest for cross-project co-ordination? - The Open-O team from China Mobile plans to demo OPEN-O and OpenBaton team indicated that they may not be bale to make it to UNH Plug-fest but plan to Demo their efforts in ETSI Plugtest in February.
    Attendees: irc+GTM

Agenda of meeting November 2, 2016  14.00 UTC /7.00 PDT Meeting #10 and summary

1, Briefing on "State of MANO architecture" and feedback listed below.

MANO TSC-Board report and feedback
State of MANO in OPNFV-tsc-borad-october-28-2016.pdf

2. Cross-Project Co-ordination for Danube and Assess the participation of MANO  activities for UNHV Plugfest Dec 5-8

3. Functest through common test plans for MANO VNFM and NFVO  (Opera, Orchestra, and participating Projects)

Added comments to JIRA ticket. Await response from Morgan if we need a new JIRA ticket or plan to persue with this for both Opera and Orchestra for VNF Blueprint testing.

https://jira.opnfv.org/secure/RapidBoard.jspa?rapidView=69&projectKey=FUNCTEST&view=planning&selectedIssue=FUNCTEST-237

4. Any topics of interests to upstream and Projects in OPNFV.

Updated the ETSI Mano page for NSD, SO/RO with acronyms in alpha order.

https://wiki.opnfv.org/display/mano/ETSI-MANO

Updated new  weekly scheduled meeting Wednesday 14 UTC for MANO WG  at top of this page.

Agenda of meeting October 24, 2016  20.30 UTC /13.30 PDT Meeting #9 and summary

  1. Inputs to TSC for Board "State of MANO architecture (Multi-modal)"
    1. Predictable - VNF Onboarding - Discussions
    2. Exploratory - Integrating MANO components in OPNFV and Hello world  Blue Prints efforts for VNF LCM.
    3. Evolving Architecture - Top Down vs. Bottom up - Prakash
    4. MANO Project Guidance
  2. Review Internal Process
    1. Need change of time (with Daylight Saving changes and to help EU/China participation)
    2. Need web page updates and review.
    3. Stick to one irc channel allocated : #opnfv-mano and help needed for taking notes

Summary:

 Actions : 1.  Prakash  will work with Bryan Sullivan to prepare a Deck to present on "State of MANO in OPNFV", at TSC/Board meeting at Barcelona.

                2. David McBride will get in touch with Orchestra PTL [Giuseppe Carella (Frauhaufer de)] to encourage them to run through the D-release plans for OPNFV integration.

                3. Opera PTL[ Yingjun Li (Huawei)] went through process to adapt, for Integrating NFV-O and VMFM over OPNFV VIM as in item a & b and the edits reflect the changes in those links.

                    Main point to note is exception for having a scenario in OPNFV to catch up on version mismatches for OPEN-O to OPNFV. This scenario may need resources for CI/CD which will be evaluated for Release D.

                    Opera PTL

                4. Kashyap from HP joined on understanding OPNFV approach to on-boarding and due to lack of time we discussed only catalog portion and that resulted in action item for mano wg as a whole..

                    Thus focus must b on what are available VNF Packaging formats and which are widely used and why as part of VNF on-boarding. CSAR link in VFN o-boarding needs to be current from OASIS.

                5. Bryan pointed the requirements brought out by upstream like OPEN-O and vendor Intel who were interested in Market Place and that's added as one item under VNF Boarding and can be edited and expanded by mano wg.

 

Attendees:

Prakash Ramchandran (Futurewei Tech. Inc)

Bryan Sullivan (AT&T)

Lawrence Lamers (VMWare)

Yingjun Li (Huawei)

David McBride (Program Manager -OPNFV/LF)

Kasyap (HPE - NFV)

 

Agenda of meeting October 8, 2016  20.30 UTC /13.30 PDT Meeting #8

Agenda:

Agenda bashing

Briefing on Open Daylight Developer Design Forum between MANO WG and Polestar (Process & work split)- Bryan

Continuing debate on on-boarding flow and processes.

The architectural issues of "catalog" amongst different VNFM, NFVO and possible options to minimize gap

action follow ups...

Summary of last meeting:

1.relationship between Polestar and MANO WG (dneary, 20:36:58)

a.Prakash reported a conversation between Margaret Chiosi, Chris Price and himself (dneary, 20:37:59)

b.Yingjun told us about VNF onboarding during Polestar meeting in ODL Summit in Seattle, but left before the end and did not see the conclusion (dneary, 20:38:44)

c.Bryan has agreed to use Jira for tracking VNF Onboarding task according to Prakash (dneary, 20:39:34)

d.conversation re VNF onboarding - 2 topics including VNF onboarding and VNF packaging (dneary, 20:48:53)

e.3rd aspect is VNF catalog (dneary, 20:49:45)

f.further discussion on how to include VNFM in scope for OPNFV - do we need VNFM scenarios? There was a suggestion to deploy the VNFM as a virtual application through Functest as an application running on the NFVi (dneary, 21:27:22)

g.Prakash is looking for a PTL for a project to start a VNF Catalog project (dneary, 21:31:59) - Could not confirm from OPEN-O as they were planning to use OPNFV artifact url as  Catalog.

Agenda of meeting Sept  21, 2016  20.30 UTC /13.30 PDT Meeting #7 (pre-ODL summit meeting) and Minutes

Circled through listed Agenda

Discussed Proposals from PowerPoint listed in agenda. 

Role of Polestar (user view/ requirements templates, concept validation, Concept Overlap)

Role of mano-wg ( assess and document best practices working with projects that work in mano stack, also share info with other working groups)

Pointed some comments in mailing about Dovetail separation

Details to be discussed at ODL Summit by Polestar / mnao-wg refer link Meetup at OpenDaylight Developer Design Forum

Questions on Specs and   Blueprints : we need terminology change in specs means OPNFV specs

Blueprint here may mean VNF descriptor, Information/Data Model, NSD , VNFFG we need some term for Blueprint to describe this precisely

 Action : Prakash to send summary though dlist.

Attendees:

Prakash Ramchandran - Futurewei Tech. Inc.

Serge Manning- Sprint

[Goal: Documenting specification of Best-practices for MANO]

Agenda of meeting Sept  7, 2016  20.30 UTC /13.30 PDT Meeting #6

Agenda Bashing

mano-wg-plan-v1.0

(JIRA tickets can be optional for Projects that want tracking)

1. VNF On-boarding scope in Polestar vs. opnfv-mano-wg (user view vs. Practice)

Please subscribe to mano-wg through this link

Models Meetings (refer October 19th/2016 notes of Bryan)

Domino Meeting Notes (refer October 19th/2016 notes of Ulas)

Projects and topics that thread through many of the systems and processes that make up a MANO stack, as above (e.g. business processes that are outside OPNFV/OpenSource scope).

VNF Packaging (Models -Pending) - Compare and contrast  in how JuJu and Tacker  use OpenStack Glance or SDN-C (ODL/ONOS/OC}  {how do they align with IFA011?}

This analysis is underway (or at least in scope) in the Models project, as a test/documentation exercise. The current state of that project is setting up the test environment (Cloudify, Tacker, JuJu, OpenBaton, etc) to assess how MANO-stack projects require/support various VNF Package elements, starting with TOSCA-based blueprints for reference VNFs (which will become part of the OPNFV CI/CD inventory along with vIMS). I welcome the support of anyone that wants to focus on the specific question Prakash mentioned.

2.Mapping between TOSCA and NFV for using VNFD from oasis for VNFM to use (http://docs.oasis-open.org/tosca/tosca-nfv/v1.0/csd03/tosca-nfv-v1.0-csd03.pdf ) - need - Arthur Tyloc & for Tacker (TBD) along with OPEN-O/OpenBaton Participation here

? Review how Service Descriptors from TOSCA match with NFV and if mapped correctly can  Template Distribution and Parsing manage this to deliver to orchestrators   like Heat or any controller that can support Pub/Sub via OPNFV project like Domino/Parser?

Bryan -> VNF Package cataloging/distribution is part of on boarding process.( with other VNF assets). The mapping of elements (e.g. Service Descriptors) is part of the information model – data model analysis that needs to be done; work that so far is still in the SDO realm AFAICT.

3. OPEN-O and use case for vCPE - (Opera)

4. OpenBaton use case vIMS/VNF On-boarding (Orchestra)

5. VES & Monitoring formats and  attributes to collect? (This work is listed in Models working with VES)

6. How do we document analysis besides Wiki page updates, any suggestions?  refer mano-wg-plan-v1.1

Please add any items you like to bring this week above as item 6...etc.

Agenda & Summary of meeting Aug 24,2016 (Toronto) Meeting #5

Includes both Toronto inputs and subsequent meeting same day:

1. Welcome to MANO WG -

Goals: A. Guide and Support for standards to adopt. MANO Stack in OPNFV &  B. Roles to ensure Service Providers and Vendors convergence on gaps to cover leveraging this WG.

2. Agenda bashing and last meetings summary

There was suggestions for improving communication and focus by  Hackfest and summarized here by Prakash.

.a. The link to meeting time and location for MANOP WG must be synched or maintain a single copy and send invites to participants via   [opnfv-tech-discuss] [manowg]

Cross Projects - Tasks that were proposed and discussed and how to collaborate through Projects

Priority is Tagging MANOWG  in OPNFV Project JIRA tickets(ask OPNFV Projects for JIRA ticket in Projects to support MANO WG participation) What will be useful, can it be targeted (Chris Price)

b. VNF On-boarding (Models, Copper, Domino, Parser) - Can we make it Policy driven and if so how? [Larry Lammers -VMWare] Modeling language agreement, Announce you capability use service like Domino to match the capabilities requested by Network for VNF Domain. Policy rule definition to support affinity/anti-affinity rule agreement using common labels

c. Project and MANO WG can look at Monitoring aspects for FCAPS, RAS (Dave Neary - Redhat) Doctor - VIM events, Triage, Integration with Monasca, (alignment across multiple projects - Infra tools, how do we go about formats)

d. Policy rule -format specify rule, type of rule, properties key/value, targets are VNF, VDU, COnnection, VL , VNFFG (CRUD node / server ) Domino parses policy and extracts and serializes the nodes and capabilities in terms of exracted labels - Ulas Kozat - Huawei

e. Fault Data Collection and agent (VES+Doctor) - Dave Neary (Red Hat) Mointtoring Infra , Event, log Management, Telemetry, Application couters/gauges, dashbaord Tools common ones like Nagio, Sensu, snap traps, Analytics engine, Panda/Spark,etcd (requirements - inputs from Operators)

f. Catalog - URI and (Murano, JuJu, Tacker, Opera, OpenBaton) - Narnder Gupta (Canonical) Requirements Collection for info for Orchestration and what is the commonality across them that OPNFV can suggest as best practice Orchestrator goes through OPNFV (have VIM, SFC,SD drivers), Can OPNFV help formalize that

g. MANO projects: Currently participating (OPNFV+Upsream)...OPEN-O (Opera),OpenBaton(Orchestra), Models, VES, Domino, JuJu, Tacker

 

4. What is the incentive for other MANO related (Projects+Upstream) : Doctor, OSM,ECOMP,...

This is covered in Summay above and no representation from Docor or OSM yet in MANO WG.

5. Cross-project Co-ordination (goes with Models Meeting Monday even weeks monthly mornings/evenings) -Summary

The key will be to get MANOWG tags

6. VNF On-boarding (goes with Models Meeting Monday even weeks monthly mornings/evenings) -Summary

This will be the first effort for MANO WG to debate and provide best practices as guidelines through  OPNFV participating cross projects.

Attendees:

Prakash Ramchandran - Futurewei Tech. Inc.

Larry Lammers - VMware

Narinder Gupta - Canonical

Ulas Kozat - Huawei

Raghavendrachari Kamsali - HPE

7. Catalog Interfaces - Demo

8. MANO WIki Page updates.based on inputs from participants.

08/10/2016 Meeting #4

Summary of  meeting (same as last meetings agenda)

Establishing an irc with meetbot #opnfv-mano - Completed

Wiki page review and update - Continues

Other items to be taken up at next meeting on Aug 24th.

Agenda

07/27/2016 Meeting #3 ( Cancelled)

1. Welcome to MANO WG formation team (5 min)

2. Agenda bashing and last meetings summary (5 min)

3.  Review the stated goals on Wiki - https://wiki.opnfv.org/display/mano (25 min)

4.  Inputs for preparation to present at subsequent TSC for providing/seeking any approvals and or guidance (15 min)

  • How to provide cross-project collaboration?  MODEL/VES/ECOMP are looking to work with Open Baton, Domino Looking to work with Use cases in OPEN-O, Edge Cloud Use cases KVM4NFV Low Latency/Multi-site  need guidance
  • Actual work (use case, gap analysis, interface requirement etc) needs to be driven and done within individual projects.

  • How to improve projects participation (OPNFV and Upstream) - Upstream has show great intrest like OPEN-O, OpenBaton

  • How to get  engagements from other projects ( Arm and Intel have both see play in VIM part of MANO Stack, we have from Arm Bob.Moankam@arm.com and possibly uri.elzur@intel.com  to join in this discussions)

5. How do we go about upcoming projects relating to MANO (5 min)

SUMMARY:

Meeting canceled due to logistics error.

07/14/2016 Meeting #2

1. Agenda Bashing

2. Discuss broad Goals -15 min

3. Review web page and see How Model Project can drive the MANO -15 min (This we could not as Bryan was not there and he discussed this with Guiseppe and have updated his input on Wiki)

4. How we engage with projects moving up the stack - Open Baton API - A point of view -15 min (API was presented and question on Endpoints, Generic VNFM vs. Vendor Specific, and ETSI standards and lack of some details like TOSCA Array for VL,VDU etc were discussed)

5. Action items and next meeting? (Doodle poll was suggested to get a fixed Semi monthly slot and improve participation)

Summary:

1.Some adjustments to 3 as Brian was not on call for Models, but he took up with Giuseppe and sent his input which is updated OPNFV Upstream Projects. 

2.Goals needed more Projects to participate internal as well upstream.before the message is sharpened.

3. First cut of Wiki was presented and Mano wiki needed changes (on-going - Mike, Bryan and Prakash to handle that)

4. As first presentation Open Baton proposed API based  testing of Interfaces and there are some variations in TOSCA NSD, VNFD, VDUl , VL etc and therefore API based testing needs Gap analysis.

5. Doodle Poll set for MANO Work Group on request basis.

Summary:

Although meeting minutes show only two projects Domino and Movie were interested more like MODEL, OPNFV (OPEN-O), Upstream Open Baton, JuJu and Tacker showed support and thus we proceeded with seeking participation.

06/30/2016

TC Minutes 20160630

06/21/2016 Meeting #1

Berlin Summit (MANO Discussions)

First discussion on MANO WorkGroup in weekly Tech meeting was based on  meeting with OPEN-O, Tacker & JOID teams and other community members and individuals interested in MANO_WG at OPNFV Berlin Summit and other OPNFV community meetings,  suggest the following as Scope for MANO_WG.  

Suggested Scope of MANO_WG in OPNFV discussed at

High level goals are

 A> Try avoid Divergence of evolving De-facto MANO WG for MANO Stack in OPNFV requirements  and ETSI MANO ISG General and evolving Normative specs.

 B> Help MANO evolution and innovation without being held hostage to slow evolving standards for OPNFV to lead the MANO integration.

 1. Scope of MANO_WG

 a) Identify from ETSI MANO ISG Specs few use cases that are most relevant to OPNFV service providers.

 (Only Service Providers participants to vote on  this and provide the IM/DM from Standards they prefer to use)

b) Core Interfaces to consider in MANO_WG to be based on API(and/or RPC messages) for identified use cases in a).

 [NFVO & VNFM Vendors to vote on this based on IM/DM provided in a) by SPs]

                I. VIM (API Produced by  VIM consumed by  NFVO & VNFM))

                II. VNFM ( API Produced by  VNMFM consumed by  NFVO & VIM )

                III. NFVO (API Consumed by NFVO from VNFM & VNFM based on use cases to be identified) 

3.  Review options for MANO system Information (IM) and Data Model(DM)  for supporting  Generic VNFM & Generic NFVO.

                - Define Generic-VNFM & Generic-NFVO based on required API/RPC and Interfaces to support standard Publicly used VNFs

                - VNFs format for provisioning  as required by VIM (Disk Image Builder/ Glance supported) and use cases agreed using  Generic VNFM

                - Any VNF Event, Notification  and Actions needed to support any Agents for VNF service required form use cases in a)

                - This will be agreed amongst all participating Vendors and SPs for MANO_WG in OPNFV

3. .  Alignment on MANO Application Information (IM) and Data Model(DM)  for supporting  accepted use cases using Generic VNFM & Generic NFVO.

                - The use case flow will be commonly accepted for all NFVO & VNFM based on ETSI ISG MANO General specs

                - Will also be verified by Test group projects to be able to test and support these use cases

                - A common reporting format will be proposed and used for all NFVO & VNFM working with OPNFV FuncTest & Yardstick  to pass testing requirements. This can also be used for Interoperability testing in Plug-fests.

4.  No item in MANO WG will be binding on any Vendor. The idea should be to facilitate options that can be practical and implementable yet allow innovation in emerging MANO stacks for OPNFV. Also anything that help Vendors to leave generic work in OPNFV the better and they can make specific variations in their Product lines that are super-set of the generic versions. 

5. This arrived MANO WG de-facto standard should be applicable in Release-D at the earliest.

Based on D Release experience OpenBaton team like to start integration from beginning of E-Release
OpenBaton will target  Doctor for E version, Needs documents ( Colorado is here http://artifacts.opnfv.org/doctor/review/23963/releasenotes/index.html)
For D-release please check with PTL Ryta Mibu ( https://wiki.opnfv.org/display/meetings/Doctor )

 

  • No labels