Wiki

• When: Every Monday 7:00-8:00 PDT/PST (now 14:00-15:00 UTC)
• Where: IRC channel: #opnfv-meeting @ Freenode
  • Freenode Web Chat; or
  • IRC Cloud
  • or any other IRC client
• See also: R1 Etherpad: https://etherpad.opnfv.org/p/octopus
• See also: R2 Etherpad: https://etherpad.opnfv.org/p/octopusR2

According to decision on April 13, the Meetings will be IRC only.

Agenda:

• Brahmaputra Milestone D
• Follow-up on proposal for release testing
• AoB

Agenda:

• Rollcall
• Agenda bashing
• Brahmaputra Release Testing

Summary:

• We discussed how we can do release testing since the proposal from the summit breakout (use fixed PODs per installer) was rejected. Chris Price said we need to test all installers on LF PODs, so we guarantee portability. Fatih and Uli will work on a proposal for the TSC, see start on etherpad: https://etherpad.opnfv.org/p/brahmaputra_release_testing

See the full meeting minutes on IRC for the details: Minutes (html) (text) (Log)

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Meeting time next week (change back from daylight saving in Europe)
• Results of Pharos questionnaire and our next steps
• Starting work on B Jira topics
• AoB

Summary:

• Pharos lab questionnaire very helpful. 2 labs already usable, more to follow soon (Huawei, ZTE, China Mobile, Orange, Dell look promising)
• Common solution for big file transfer to China would be helpful

See the full meeting minutes on IRC for the details: Minutes (html) (text) (Log)

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Reconfiguration of LF POD 1
• Starting work on B Jira topics
• AoB

Summary:

• Lab related actions are still open. Need higher priority, since could become blocking issues for Brahmaputra.
• Agreed to continue with original LF POD1 and break it down, converting to standalone servers, then split the rest out into individual broadcast domains. That way we will have 3 PODs for bare metal (LF POD2, intel POD2, Ericsson) and several servers to run virtual PODs.

See the full meeting minutes on IRC for the details: Minutes (html) (text) (Log)

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Milestone C for B-Release
• Arno SR1 work
• Current topics:
  • Connecting ZTE lab
  • CI for armband
  • Patch for OCTO-125
• AoB

Summary:

• Cleanup of committer list progressing. Three committers agreed to step down, two committers cannot be reached. Team agreed to approach TSC for committer removal.
• Milestone C reached
• Lab setup (Arm, ZTE) is progressing

See the full meeting minutes on IRC for the details: Minutes (html) (text) (Log)

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Lab Compliancy
• Status C-milestone
• B-Release Work, use etherpad: https://etherpad.opnfv.org/p/octopusR2
• SR1 work
• AoB

Summary:

• work with Pharos about lab usage is ongoing.
• discussion on KVM4NFV workflow and CI
• Preparing the work on putting R2 backlog on Jira

See the full meeting minutes on IRC for the details: Minutes (html) (text) (Log)

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Lab Compliancy
• B-Release Work, use etherpad: https://etherpad.opnfv.org/p/octopusR2
• AoB

Summary:

• Pharos Status/Lab Compliancy
  • There have been discussions regarding the slow progress with Pharos.
  • Uli got an action to raise this to TSC and right after this Trevor sent mail to restart Pharos meetings so we should see some movement in Pharos.
  • Octopus requirements on Lab Compliancy are ready to be shared.
• B-Release Work
  • Different projects started asking questions regarding how to get their projects into CI.
  • Help and guides provided to them and waiting for further questions.
• Arno SR1
  • There have been some questions regarding where the Arno SR1 would come out from.
  • It has been clarified that the Arno SR1 will come out from stable/arno branch.

See the full meeting minutes on IRC for the details: Minutes (html) (text) (Log)

No meeting was held due to internet issues.

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Lab Compliancy
• B-Release Work

Summary:

• Action Items
  • No action items closed.
• Lab compliancy
  • Will bring up the Octopus requirements to Pharos (https://etherpad.opnfv.org/p/octopusR2)
• B-Release Work
  • Dependency towards Releng for Lab Reconfiguration

See the full meeting minutes on IRC for the details: Minutes (html) (text) (Log)

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Lab Compliancy
• Define what is E2E for CI
• Jira review and assign important topics
• LF Lab reconfiguration status

Summary:

• Action Items
  • No action items closed.
• Lab compliancy
  • Discussion is going on on etherpad (https://etherpad.opnfv.org/p/octopusR2)
• Discussions regarding E2E CI
  • See the discussion on full meeting minutes
• LF Lab Reconfiguration Status
  • Fuel adaptations/verification is going on on POD2.

See the full meeting minutes on IRC for the details: Minutes (html) (text) (Log)

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Starting clean-up of committer list
• Lab Compliancy
• Define what is E2E for CI
• Jira review and assign important topics
• Starting work on B-Release, use etherpad: https://etherpad.opnfv.org/p/octopusR2
• AoB

Summary:

• Action Items
  • No action items closed.
• Lab compliancy
  • Meimei put input to Etherpad regarding lab compliancy (https://etherpad.opnfv.org/p/octopusR2)
• Discussions regarding E2E CI

See the full meeting minutes on IRC for the details: Minutes (html) (text) (Log)

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• AoB

Summary:

• Action Items
  • Closed: Naming scheme.
    new: Chigang to try the naming scheme on a POD to see whether we need a change of token.
  • Closed: Pharos interworking.
    Pharos will provide compliance spec.
    Agreed that when labs claim to be Pharos compliant, octopus will validate that.
    new action to contact Pharos and offer some help with compliance spec.
  • Closed: Zuul demo.
    demo is recorded and available at: https://www.youtube.com/watch?v=fOYdsNdkCuw
  • Closed: Etherpad updated
    trozet, pbandzi, aricg are working on automatic configuration scripts under OCTO-109 which is needed for sequential run of CI on a single POD.
• Uli reported from internal discussions about more flexible POD usage and CI pipeline.
• AoB

See the full meeting minutes on IRC for the details: Minutes (html) (text) (Log)

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Jira review and assign important topics
• Starting work on B-Release, use etherpad: https://etherpad.opnfv.org/p/octopusR2
• AoB

Summary:

• Action Items
  • Closed: 1 (RDO/TripleO demo)
  • Open: 1 old and 4 new action items
  • See the meeting minutes for the details
• Jira Issues for Arno Maintenance
  • Reminded everyone to take a look at open Jira issues for Arno Maintenance - https://jira.opnfv.org/issues/?filter=10200
• Work on B-Release
  • Prioritization of ideas and creation of corresponding stories in Jira are needed.
  • Initial prioritization is done on Etherpad under the heading Prioritization of Ideas: https://etherpad.opnfv.org/p/octopusR2
• AoB
  • Slave naming scheme
  • Congress work
  • See the meeting minutes for the details

Full minutes see IRC: Minutes (html) (text) (Log)

Agenda:

• RDO / Triple-O demo

The meeting was done using Bluejeans to allow the screensharing.
Thanks Tim and Dan for this!

Here's the recording: https://bluejeans.com/s/8o1U/
Also here's a pdf of the slides: https://radez.fedorapeople.org/ProjectApex-RDOManager.pdf

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Feedback Arno maintenance proposal? https://wiki.opnfv.org/releases/stablebranch
• Proposal for Releng project
• Starting work on B-Release, use etherpad: https://etherpad.opnfv.org/p/octopusR2
• AoB

Summary:

• We will have the RDO/Triple-O demo on July 6.
• Fatih explained how the new releng project and Octopus will work together.

Full minutes see IRC: Minutes (html) (text) (Log)

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Arno maintenance - High prio Jira issues: https://jira.opnfv.org/issues/?filter=10200
• Starting work on B-Release
• AoB

Summary

• Action item review
  • Only action item that is discussed was RDO/TripleO demo. Preliminary date for demo is July 6th.
  • Other action items skipped due to not having acation item owners in the meeting.
• Walked through maintenance work that is still going on. There are still unfinished items.
• Slightly touched upon R2 work.

Participants: Fatih Degirmenci, Tim Rozet, Larry Lamers, Narinder Gupta, Dave Neary, Prakash Ramchandran.

Full minutes are not available due to not having meetbot functional.

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Arno CI quick retrospective - what was good, what we can improve, what was not so good
• Arno maintenance - High prio Jira issues: https://jira.opnfv.org/issues/?filter=10200
• Starting work on B-Release
• AoB

Summary

• Arno CI retrospective showed that Octopus team needs to increase.
• Planned demos for zuul and RDO/TripleO to improve R2 CI
• R2 planning will start next meeting based on etherpad

Full minutes see IRC: Minutes (html) (text) (Log)

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Feedback from Vancouver OpenStack summit
• Arno Release Finalization
• Prepare Maintenance for Arno Release
• Starting work on B-Release
• AoB

Summary:

• Discussion regarding how to ship Arno
  • Manual release process
  • Tag/branch identification and proposals to TSC

Full minutes see IRC: Minutes (html) (text) (Log)

Summary

• No attendee - no meeting.

Full minutes see IRC: Minutes (html) (text) (Log)

Agenda:

• Rollcall
• Agenda bashing
• Action item review
• Arno Release follow-up
• Maintenance for Arno Release
• Vancouver OpenStack summit
• next meeting

Summary

• Discussion whether we should add more documents to Arno release
• Discussion on the changes needed in Foreman pipeline due to the reconfiguration in LF lab
  Foreman deploy can be triggered again when scripts are adapted.
• No separate octopus session in Vancouver, but Fatih, Uli, Daniel will be there and meet informally
• Maintenance for Arno needs to be discussed. Input from https://wiki.opnfv.org/get_started/release_and_maintenance

Full minutes see IRC: Minutes (html) (text) (Log)

Actions:

• ulik and vlaza: add a note how to comment on documents under "document tracking"
• trozet for inventory update (already done)

Agenda

• Rollcall
• Agenda bashing
• Action item review
• Arno Release follow-up
• Status lab connections
• Documentation Status
• Release 2 Brainstorming

Summary

• Good progress on Documents
• Discussion whether Octopus documents should be part of the Arno release
• Not much progress for Arno, but jenkins automation for Foreman track about to start
• Now 7 labs are connected, soon 8.

Full minutes see IRC: Minutes (html) (text) (Log)

Actions

• ulik and vlaza: add a note how to comment on documents under "document tracking"
• ulik: ask around… and then let's come back with the question whether we need something of octopus in arno docs.

Agenda

• Rollcall
• Agenda bashing
• Arno Release follow-up
• Move main wiki pages to git
• Release 2 Brainstorming

Summary

• Deployment in Ericsson and Intel labs are "running" according to BGS, but manual baremetal deployment on LF lab still being worked on.
• Some trials on test done.
• Documentation on pipeline and slave connection documentation waiting to be merged. Please review.
• Main wiki pages related to releases should be on git.
• Some actions defined to achieve this.

Full minutes see IRC: Minutes (html) (text) (Log)

Agenda

• Status of LF hardware
• Arno-related Jira topics
  • https://jira.opnfv.org/issues/?jql=project%20%3D%20OCTO%20AND%20status%20in%20%28Open%2C%20%22In%20Progress%22%2C%20Reopened%29%20AND%20labels%20%3D%20R1
• Documentation status
• Review process

Summary

• Arno related topics are now labeled "R1". Let's be very clear with information in Jira to help Arno planning.
• Good progress on documents
• Decided to clearly follow the process as distributed in email:
• Good experience with IRC-only, so let's stay this way.

Full minutes see IRC: Minutes (html) (text) (Log)

Agenda:

1. Welcome to new committers
2. Meeting format (use of IRC)
3. Status update: connecting labs
4. Status update: artifact repository
5. Achieving RC2 (Automatic testing on LF hardware fully functional)
6. Octopus documentation for Arno
7. First thoughts about Octopus in Release 2
8. AoB

Summary:

1. Octopus is well prepared for RC2, but needs about 1 day of work after LF hardware becomes available (that is connected to CI via its jenkins slaves.
2. Agreed to prepare Octopus documentation to be included in Arno release
3. Starting a brainstorming for release 2 tasks in Octopus.
4. Next meeting will be IRC only

Full minutes see IRC: Minutes (html) (text) (Log)

1. Committer promotions
2. Triggering tests in different labs
3. Using hardware/VMs/Docker
4. Artifact repository

1. Committer promotions:
   Fatih, Iben and Jerry promoted as committers to Octopus
2. Release candidate 1:
   RC1 only needs build running, which is currently the case, but only with Ericsson labs.
   Also build scripts for other installers missing.
3. Triggering different labs:
   There are difficulties to connect intel lab. It will be good to work in parallel to connect different labs and learn from each other.
4. Using hardware:
   In Arno we will just use bare metal. Build/deploy jobs in VMs/Containers we will revisit for R2.
5. Document repository:
   Text based systems like markdown or latex are better to use with git, since they allow change based reviews. Therefore we will recommend to TSC to use such a system for all. Octopus own documentation for Arno shall be done in markdown.
6. Artifact repository:
   Access problem seems to be solved.
   Usage of metadata needs more clarification.

Minutes (html) (text) (Log)

• Recap from Test&Performance meeting
• Relation between Functest, BGS and us
• How to trigger the functest (connecting the lab servers)
• Automation
• Start to define the way we are using our hardware
  VMs (or container) vs. bare metal
  Which jobs should run virtualized?
  How many jobs running in parallel? etc.
  We might need some input from BGS here.
• Artifact repository
  See https://wiki.opnfv.org/wiki/artifactrepos

• Connection between LF and other labs to be implemented using JNLP, as described on etherpad (https://etherpad.opnfv.org/p/octopus).
  Ericsson lab is already connected.
• Different ways to use hardware discussed, no conclusion yet.
• There seems to be difficulties to access Google storage from China, so discussing other solutions.

Minutes (html) (text) (Log)

Discuss octopus requirements on the scripts in project repos that will be called by Jenkins.
(let’s compare these also with the build script currently in review at https://gerrit.opnfv.org/gerrit/144 )

Minutes (html) (text) (Log)

1. Walk through CI workflow with specific emphasis on usage of the tools incl. jjb
2. See currently working jobs as examples
3. Identify the tasks that can be distributed to all contributors/committers in the octopus team
4. Other topics

Minutes (html) (text) (Log)

Minutes (html) (text) (Log)

please see hackfest minutes

Agenda

1. Experience and status of setup tools in LF – Fatih
2. Solutions for permissions for CI – Jerry (Zhaoxinyu)
3. Discuss proposals on etherpad https://etherpad.opnfv.org/p/octopus and try to work out a timeline for Rel 1 parts

Minutes

• Fatih explains his work on the LF setup and walks us through the new infrastructure wiki page:
  https://wiki.opnfv.org/octopus/infrastructure
  We discuss plugins and jenkins permissions
  It was agreed to use jenkins job builder, so not everybody creating a job needs jenkins permissions.
  
  Uli will ask LF to provide permission to Fatih and Jerry and install the cron job job builder application and necessary plugins for this.

• Fatih explains his proposal for release 1: https://wiki.opnfv.org/octopus/flows
  The page talks about contributors/committers of other projects in that flow. The first commit should be getting BGS output to rerun BGS work automatically.
  Frank confirms that this is the goal.
  
  All should review this wiki page and further define these work flows.

• Short discussion on test inputs. We will use tempest, robot and/or rally, but functest project should provide the details.

MeetBot Minutes

http://meetbot.opnfv.org/meetings/opnfv-meeting/2015/opnfv-meeting.2015-02-16-15.13.html

Agenda

1. Contact person for LF and my application for committer – Uli
2. ARM (not the company, but Artifact Repository Manager ) – Fatih, Iben
3. Jenkins Slaves – where to run them – Fatih, Aric, Iben
4. Starting to design the CI framework – Lijun

Minutes

• Uli asked whether the project agrees that he will be the contact person for Linux Foundation. People agreed. Several committers already have voted per email that Uli should be also accepted committer priviledge. So Uli will ask the TSC to approve this and notify LF.
• We discussed the necessity of a repository to store the artifacts, since they shouldn't go into git. Iben Rodriguez will ask LF which ARM they are using in other projects. Probably the same ones would do for us.
  Prakash proposed also to consider swift. We probably need local and central repos depending on the kind of workflow (developer using CI or periodic automatic builds).(See here.)
• We also discussed limits of file size for git. The general agreement was to start with lower limits and increase if necessary. We can use LF's suggestions for initial limits. (See here.)
• Jenkins slaves do background processing of raw log files and measurements and create visualization. Fatih clarifies that not only this but also the main tasks we need to decide where to run.
  Iben offered to provide background information how to setup jenkins on the wiki (as well as other tools). Please provide feedback to this wiki contents where ever things are not clear. (See here.)
• We discussed usage of IRC in parallel for our meetings and agreed to use one common IRC channel (#opnfv-meeting). This helps to avoid overlapping meeting problems. (See here.)
• We will include information about octopus activities on the upcoming two hackfest/meetups on wiki. Please contact Prakash for Santa Rosa or Uli for Prague questions. Please see and update Etherpad from link: (See here.)
• Last topic of agenda couldn't be addresses because time was up.

Agenda

1. Follow-up on discussion on relation between Octopus and other OPNFV projects, feedback from TSC
2. First ideas on bottom-up thinking, tools, inputs from BGS work
3. Updating Octopus web page

Minutes

We used the updated diagram for the discussion.
(see slideset octopus_diagram_flow_v4.pptx or source octopus_diagram_v4.rar.

The diagram shows, how Requirement Projects and Collaborative Development Projects provide inputs to upstream projects. Octopus from there gets the repositories and tools.
Integration projects like BGS investigate the detailed steps how to build the OPNFV platform from a certain set of upstream projects and tools. This information is provided to Octopus via a cookbook. The cookbook defines the exact steps to execute in the grey box (here as an example shown as "Build-Verify-Package/Image").
Testing projects will define how to do verification in CI and provide inputs to the deployment and the testcases.
Octopus feeds some packages/images (format and content needs to be defined) to the deployment.
Different paths in the deployment will be necessary and might also be specified in the cookbook.

The second slide in the Powerpoint shows a workflow for a collaborative development project providing a patch to some upstream project that is then through Octopus CI and Deployment provided to the test lab.

Questions discussed:

• Relation BGS-Octopus
• Which kind of information do we need in the cookbook
• we should refine requirements to other projects
• we need to be able to define different configurations.
  BGS only works for a 5-node configuration, while Copper currently
  investigates a 1-node configuration. How does that affect the
  CI-Output? Will it be part of the cookbook or can that be independent?

Next steps:

• Prakash will start an etherpad to follow up on these questions
• Fatih will work on first descriptions.

Note: here is the etherpad: https://etherpad.opnfv.org/p/octopus

Agenda

1. Follow-up on last week's discussion on relation between Octopus and other OPNFV projects
   see diagram
2. Let’s look at the BGS time-line and its implications for our project
3. Start discussing what should be our deliverable in release 1, since there will be no full CI process needed in that timeframe
4. If possible get some more input by Dan about RDO and the tools provided from there for integration.

Minutes

• The diagram prepared by Fatih and Uli reflects the common understanding. We will put this on the main Octopus wiki page so everybody in OPNFV can see how the general workflows with CI should work
  All team members are requested to review and improve this description.
• When looking at BGS timeline, Octopus can provide to release 1 the ability to run what BGS specifies in the cookbook in a CI flow. That is starting from BGS M3 (Complete Alpha Installation Package), we can work on this cookbook to be able to repeat what it specifies. If possible, also a M4 based cookbook could be run.
• Dan provided a status update of his work about the usage of puppet modules for installation of OpenStack with ODL. This is very similar to what BGS does, but already includes jenkins. He plans to show this on the meetup in Santa Rosa end of february.
• Next meeting next Monday same time.

Agenda

1. I like to discuss some initial material we have prepared about possible next steps
2. Let’s look at the BGS time-line and its implications for our project
3. Let’s look at OSCAR’s diagram of the flow from BGS to CI and CD and discuss its implications for CI
4. Start discussing what should be our deliverable in release 1, since there will be no full CI process needed in that timeframe

Minutes:

• Prakash explained the OpenStack and ODL way of doing CI (see slides)
• We quickly saw we need to understand the relation of Octopus to the other projects. So we looked at the Oscar diagram.
• Fatih Degirmenci will together with myself work on a diagram of embedding Octopus between BGS (and other inputs) and OSCAR
• There were a few voices promoting the Openstack way of doing CI.
• Dan explained the way, RDO combines the various tools for integration. Dan will come with more inputs for the next meeting.
• We were not able to look at the timeline of BGS inputs and discuss what we can do for Release 1. We have to do that next week.

Agenda

1. Welcome (Uli)
2. Short introduction round (all)
3. Communication methods, project infrastructure etc. (will ask Ray/Ariq)
4. First steps/tasks proposal (ChrisW)

1. Inputs for the discussion on the detailed task list were not yet available, so we postponed the topic.
2. We now have a good list of contributors and committers that should be fine for the approval by TSC.
3. Uli reported Palani's request to have a close colaboration with the OPNFV testing project (OPNFV - Base system functionality testing)
   and encouraged the team members to contribute also to this project. Octopus is depending on that project because it needs to use the test objects created there.
4. For the start-up of the project, weekly meetings are envisaged.

1. Inputs for the discussion on the detailed task list were not yet available, so we postponed the topic.
2. Contributors / Committers / Maintainers
   Frank explained the responsibilities of committers, but it is unclear what the role of maintainer means.
   The project proposal template mentions this role.
   We should have a couple of committers defined soon. All volunteers so far will be listed in the “contributor” section.
   AI Uli: update proposal accordingly
   AI Uli: send email to all contributors and ask them about their long-term availability for a committer role.
   AI Uli: clarify with TSC about the role of maintainer.
3. Olivier Jacques presented the forj project (a tool to replicate OpenStack’s CI/CD infra anywhere) which could be of help for octopus.
   The slides and other information can be found on www.forj.io.
   We need to check whether this helps also in relation to Linux Foundation, since they already provide us with git, Jenkins, etc. environment.
   AI Uli: Ask Ariq and connect him with Olivier
4. Next meeting next Monday, same time
   Uli will put the meeting details on the wiki in octopus page and meeting page and send an outlook invite to all contributors to the project.

• We did some minor online changes to the project proposal.

• Frank noted, that is still not clear what are the exact actions to be done to start the work in the project. We then tried to understand how this is tried to be covered in the 4 bullets in the summary section.

• Chris W offered to prepare better description together with Dan describing the pipeline from build/patch build to deployment and testing.

• We discussed that octopus should provice “patchability” for major upstream projects. But it is difficult to identify at this time, for which projects this will be needed besides OpenStack.

• We need to clearly describe which of the steps in octopus need to be done for use in the first release.

• We will continue to work on this via email and wiki editing and have a follow-up meeting in a similar way after Thanksgiving.

• Main results are captured in a word document as comments or change marks.

• We had a very long discussion about the level of detail we need in the document.

• We concluded that some information about the first step would be very helpful for people providing resources.

• We try to address this with some additions which will be done in the wiki during the next 2-3 days.