SuccessBot Says

• Bknudson: we got rid of keystone CLI [in favor of OpenStack Client]
• jrichli: it has been shown that Swift encryption can pass all functional tests.
• Bauzas: only a very few Nova changes were missing a reno file, the team is now super-trained on getting them.
• Odyssey4me: OpenStack-Ansible now has a Designate role ready for testing [1].
• ttx: Glance is the first project to issue RC1!
• Mugsie: mlavalle completed the Nova/Neutron/Designate DNS Integration along with docs + clients.
• Odyssey4me: OpenStack-Ansible has released Kilo 11.2.11. It’s the first time that we’ve used the release team for a release and we love it!
• Odyssey4me: OpenStack-Ansible Liberty 12.0.8 has been released.
• Tell us yours via IRC with a message “#success [insert success]”.
• All Successes

Current PTL Election Status

• Important dates:
  • Election open: 2016-03-18 00:00 UTC
  • Election close: 2016-03-24 23:59 UTC
• Projects with only one candidate: 41
• Projects with no PTL candidates:
  • EC2-API
  • Stable Branch Maintenance
  • Winstackers
• The TC will appoint a new PTL for projects without a candidate [2]
• Confirmed Candidates [3]

Quotas – Service vs. Library

• There is a spec for cross-project Quota work [4] that is seeking feedback to move ahead as a service or library.
• Service:
  • New project to manage quotas for all projects that use the service.
  • It will be responsible for handling the enforcement, management and DB upgrades of the quotas logic for all.
  • However, all projects would have a big dependency on this one service.
• Library – two ways:
  • Does not deal with database models
    • Maybe a ABC or even a few standard implementation vectors that can be imported into a project space.
    • The project will have it’s own API for quotas and the drivers will enforce different types (e.g. flat quota driver or hierarchical quota driver) with custom/project
    • Project maintains it’s own DB and upgrades.
  • A library that has models for DB tables that the project can import from.
    • Projects will have a handy outline of what the tables should look like.
    • Project has it’s own API and implements drivers in-tree by importing this semi-defined structure.
    • Project maintains it’s own upgrades but will be somewhat influenced by the common repo.
• Or just avoid all this simply give guidelines.
• A service has been proposed in the past with projects like Boson [5].
• Tim Bell raises initially a library would be good.
  • If we can’t agree on a library, we’re unlikely to agree on a service.
  • Would allow for consistent implementation of nested and user quotas.
• For projects like Trove that need a consistent lock on quotas of all projects, there are race condition issues for projects like Nova that need to be solved first.
• The main issue with doing a library that was raised in a previous summit was how to tie in database table management with the existing tables owned by a project. While this is not impossible to solve, we need to think about which tools can help with that.
• Full thread

SuccessBot Says

• Ajaeger: All jobs moved from bare-trusty to ubuntu-trusty.
• Clarkb: Infra is running logstash 2.0 now
• Tell us yours via IRC with a message “#success [insert success]”.
• All Successes

Cross-Project

• API guidelines ready for review:
  • Header non proliferation [1]
  • Client interaction guideline for microversions [2]

Election Season, PTL and TC

• PTL elections:
  • Important dates:
    • Nominations open: 2016-03-11 00:00 UTC
    • Nominations close: 2016-03-17 23:59 UTC
    • Election open: 2016-03-18 00:00 UTC
    • Election close: 2016-03-24 23:59 UTC
  • Every project team must elect a PTL every 6 months.
  • More info and how to submit your candidacy [3].
• TC elections:
  • Important dates:
    • Nominations open: 2016-03-25 00:00 UTC
    • Nominations close: 2016-03-31 23:59 UTC
    • Election open: 2016-04-01 00:00 UTC
    • Election close: 2016-04-07 23:59 UTC
  • Under the rules of the TC charter [4] we renew 7 TC seats. Seats are valid for one year.
  • More info and how to submit your candidacy [5].
• Full thread

The stable:follows-policy Tag Is Official, Projects Need To Start Applying For It

• This is official in the governance documents [6].
• Projects that follow the stable branch policy [7] should start applying.
• Full thread

Release Countdown For Week R-3, March 14-18

• Focus:
  • Projects teams following cycle-with-milestone model:
    • Preparing their first Mitaka release candidate this week.
    • This should be tagged using (X.Y.Z.0rc1) as soon as the pressure to unfreeze master is stronger than the cost of backporting bugfixes.
    • The release team will create stable branches from the release candidate tag points.
  • Project teams following the cycle-with-intermediary model
    • Ensure you have at least one mitaka release.
    • Determine if you need another release before the end of the Mitaka cycle.
  • All feature freeze exceptions that haven’t landed at this point should wait until Newton.
• General Notes:
  • The global requirements list is frozen. If you need to change a dependency, for a bug fix, please provide enough detail in the change request to allow the requirements review team to evaluate the change.
  • User-facing strings are frozen to allow the translation team time to finish their work.
• Release Actions:
  • The release team has started creating the stable/mitaka branches for libraries.
  • Follow-up on the mailing list thread [8] to acknowledge and approve the version number to use to create the branch.
    • This only includes projects with release:managed tag.
    • Other projects can post on the thread of request their own branches.
• Important Dates:
  • RC target week: R-1, March 28 – April 1
  • Mitaka final release: April 4-8
  • Mitaka release schedule [9].
• Full thread

Reminder: WSME Is Not Being Actively Maintained

• Chris Dent and Lucas Gomes have been actively verifying bug fixes and keeping things going with WSME, but are no longer interested or have the time to continue. It was also found it never really reached a state where any of the following are true:
  • The WSME code is easy to understand and maintain.
  • WSME provides correct handling of HTTP (notably response status and headers).
  • WSME has an architecture that is suitable for creating modern Python-based web applications.
• There’s a suggestion for the 24 different OpenStack projects that are using it to move to something else.
• One big reason for choosing WSME earlier was that it had support for both XML and JSON without application code needing to do anything explicitly.
  • The community has decided to stop providing XML API support and some other tools have been used instead to provide parsing and validation features similar to WSME:
    • JSONSchema
    • Voluptuous
• Full thread

SuccessBot Says

• ttx: Mitaka-3 is done.
• Odyssey4me: OpenStack-Ansible Liberty 12.0.7 is released [1].
• johnthetubaguy: Nova is down to four pending blueprints for feature freeze now [2], sort of one day left. Better than it was this morning at least.
• Russellb: Got a set of OVS flows working in OVN that applies security group changes immediately to existing connections.
• Tell us yours via IRC with a message “#success [insert success]”.
• All

Cross-Project

• Quotas and Nested Quotas Working group
  • Meeting [3]
  • Spec [4]

Outreachy May-Aug 2016: Call For Funding and Mentors

• Outreachy [5] helps people from groups underrepresented in free and open source software get involved by matching interns with established mentors in the upstream community.
• We have 10 volunteer mentors for OpenStack this next cycle (May 23-August 23 2016).
  • Learn more and apply to be a mentor [6]
• Potential sponsors have reached out, but we need more due to the increase in applicants.
  • Each intern is $6,500 for the three-month program.
  • The OpenStack Foundation has confirmed participation.
  • Learn more and apply to be a sponsor [7].
• Regardless, help spread the world!
• Full thread

Changing Microversion Headers

• The API working group would like to change the format of headers used for microversions to make them more future proof before too many projects are using them.
  • Proposed guideline [8].
• This came up in another guide for header non-proliferation [9].
• After plenty of discussions, and with projects already deploying microversions (Nova, Ironic, Manila) the proposal is change basic from:
  • X-OpenStack-Nova-API-Version: 2.11
  • OpenStack-Compute-API-Version: 2.11
• To:
  • OpenStack-API-Version: compute 2.11
• This allows us to use one header name for multipel services and avoids some of the problems described in the header non-proliferation guideline [9].
• Full thread

OpenStack Contributor Awards

• The Foundation would like introduce some informal quirky awards to recognize the extremely valuable work that we all do to make Openstack excel.
• With many different areas to celebrate, there are a few main chunks of the community that need a little love:
  • Those who might not be aware that they are valued, particularly new contributors
  • Those who are the active glue that binds the community together
  • Those who share their hard-earned knowledge with others and mentor
  • Those who challenge assumptions, and make us think
• Nominate someone who you think is deserving of an awards [10]!
• Full thread

Status Of Python 3 In OpenStack Mitaka

• 13 services were ported to Python 3 during the Mitaka cycle: Cinder, Glance, Heat, Horizon, etc.
• 9 services still need to be ported
• Next Milestone: Functional and integration tests
• “Ported to Python 3” means that all unit tests pass on Python 3.4 which is verified by a voting gate job. It is not enough to run applications in production with Python 3. Integration and functional tests are not run on Python 3 yet.
• Read the full status post [11] by Victor Stinner.
• Join Freenode channel #openstack-python3 to discuss and help out!
• Full thread

Audience for Release Notes

• We have 3 potential audiences for release notes:
  • Developers consuming libraries or other code directly.
  • Deployers and operators.
  • End-users.
• Two kinds of documentation being discussed:
  • Reno [1] for release notes [2]
    • Keep in mind the audience is *not* someone who is necessarily going to be looking at code or writing apps based on libraries we produce..
    • Highlight information that deployers and operators will need, like changes to configuration options or service behaviors.
    • Describe REST API changes that an end-user may need to know about.
  • In-tree developer documentation [3] for Developers.
    • Internal details, library API changes, etc. — any changes the deployer is not going to see.
• The two sets of documentation are meant for different purposes, so we need to think about what information we publish in each.
• Full thread [4]

A Proposal to Separate the Design Summit

Today

• Since the beginning, we’ve had face to face time at Design Summits to discuss development cycles to set goals and organize the work for the upcoming 6 months.
• A more traditional conference took place at the same time to ensure interaction between upstream and downstream parts of our community.

Current Issues

• For developers:
  • Difficult to focus on upstream work with so many distractions coming from the conference.
  • As a result the summit is less productive, and we form midcycles to fill our focus.
• For companies:
  • Flying all their developers to expensive cities and conference hotels for the summit is pretty costly. – The goals of the summit location reaching out to users everywhere does not necessarily align with the goals of the design summit location.
  • Not enough time to build products on top of the recent release.

• Not enough time for users to try out the new release to have feedback.
• Finding venues that can accommodate both events is becoming increasingly complicated.

How to Split up the events

• The first event would be for upstream technical contributors of OpenStack.
  • Held in a simpler, scaled-back setting that would let all OpenStack project teams meet in separate rooms, but in a co-located event that would make it easy to have ad-hoc cross-project discussions.
  • It would be held in closer to the centers of mass of contributors, in less-expensive locations.
  • It would happen a couple of weeks /before/ the previous cycle release. There is a lot of overlap between cycles. Work on a cycle starts at the previous cycle feature freeze, while there is still 5 weeks to go. Most people switch full-time to the next cycle by RC1. Organizing the event just after that time lets us organize the work and kickstart the new cycle at the best moment. It also allows us to use our time together to quickly address last-minute release-critical issues if such issues arise.
• The second event would be the main downstream business conference.
  • This includes high-end keynotes, marketplace, and break out sessions.
  • Organized two or three months /after/ the release, to give time downstream users to deploy and build products on top of the release.
  • This will better allow us to gather feedback on the recent release, a gather requirements for the next cycle.
  • A subset of contributors who would like to participate in sessions can collect and relay feedback to other team members (similar to the operators midcycle meetups).
• The split should reduce the number of midcycle events, however, if they’re still needed, they could happen at the conference event (which is in the middle of the cycle).
• The split means that we need to stagger the events and cycles. We have a long time between Barcelona and the Q1 Summit in the US, so the idea would be to use that long period to insert a smaller cycle (Ocata) with a release early March, 2017 and have the first specific contributors event at the start of the P cycle, mid-February, 2017. With the already-planned events in 2016 and 2017 it is the earliest we can make the transition. We’d have a last, scaled-down design summit in Barcelona to plan the shorter cycle.
• Operators midcycle will still continue to happen.

Voiced Concerns and Answers:

• This creates two events instead of one. Creates community split, with developers skipping the main and non-developers not providing any feedback to the contributor specific event.
  • There will still be a lot of strategic discussions at the main event.
  • This is where we look at the N-1 release and start drawing plans and cross-project themes for the N+1 release.
  • We don’t need every developer there, but we still need a significant chunk of them, with every team represented, so that we can have those strategic and cross-project discussions.
  • Someone who wants to keep touch with development could still make only one trip, so it’s not expected for the communities to be split. We’d still all be represented in the main event.
• Losing the main summit as an excuse to fund developers’ travel. Some developers are only sent to the Design summit because the main event is happening at the same time.
  • If you have to pretend to attend the Summit to be able to attend the Design Summit instead, there is deception involved. You should a talk with your employer on where the most value lies for you in attending which event.
  • The Foundation also has the Travel support program to cover the gaps [5].
• The fear of US-centricity (translating from “closer to the centers of mass of contributors”). This makes travel cost cheaper at the expense of making it more costly for non-US parts of our community.
  • The goal is “minimize and *balance* travel costs for existing contributors”. There will still be some continent rotation involved, but we need to balance cost and fair.
• The lost of midcycle spirit. Some people really like the midcycles as they stand: separated small events where only your small team is around. The split appears to reduce the likelihood, the need, or the funding for such events.
  • While the hope is the proposed format will let us fulfill all our team socialization needs, it’s true that there will be other people around, and it will feel a lot less exclusive and special.
  • The trade-off is that having people all together encourages cross-project work and breaks silos.
• Full thread [6]

OpenStack Operators

• Takeaways from the OpenStack Mid-Cycle Ops Meetup: first time’s the charm [1]
• OpenStack Upgrading Tutorial: 11 pitfalls and solutions [2]

“No Open Core” in 2016 (continuing)

• Continuing with discussions on Poppy in particular, Doug Hellmann raises that the Poppy team has done everything we’ve asked. The governance currently emphasizes on the social aspects of the project and community interactions. Tell the Poppy team they “are not OpenStack” even though they followed things is distressing.
• Sean Dague mentions that solutions like Neutron have an open source solution. It may needed some work, but at least there was one open source solution for testing.
• Dean Troyer brings up how we have things like Cinder with commerical products as storage drivers. Even without those storage drivers though, Cinder is still useful.
• Poppy’s open source implementation option, OpenCDN is currently an abandoned project.
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-February/085855.html

Upgrade Implications of Lots of Content in paste.ini

• A set of Cross-origin resource sharing (CORS) patches came out recently [3], that adds a ton of content to paste.ini.
• Paste.ini is a config that operators can change.
  • Large amounts of complicated things in there which may change in future releases is really problematic.
  • Deprecating content also is a challenge.
  • Why aren’t these options just the defaults of the CORS middlware?
    • Some projects like Ironic did request to not have headers prescribed, for eample, if Ironic is using something other than Keystone, different allowed headers would be required.
    • However, Keystone is defcore, so the default should be useful there first. Then be flexible to other auth options can go on top.
• Where do we go from here?
  • Option 1: Implement as is, keep things consistent, fix them in Newton.
    • This is not fixable in Newton as it requires deprecating out for the next three releases.
  • Option 2: Try to fix it in Mitaka 2 of CORS middleware setting defaults and projects having the ability to override [4].
    • This will require patches against a bunch of projects [5]. Who can help?
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-February/086746.html

[1] – http://superuser.openstack.org/articles/takeaways-from-the-openstack-mid-cycle-ops-meetup-first-time-s-the-charm

[2] – http://superuser.openstack.org/articles/openstack-upgrading-tutorial-11-pitfalls-and-solutions

[3] – https://review.openstack.org/#/c/265415/1

[4] – http://docs.openstack.org/developer/oslo.config/generator.html#modifying-defaults-from-other-namespaces

[5] – http://governance.openstack.org/reference/projects/index.html

SuccessBot Says

• dhellmann: [1] is live with release history and upcoming release schedules.
• ttx:  Governance changes are now announced in #openstack-dev, to increase general awareness about them!
• jklare: Just merged the first batch of opentack-chef cookbook patches for the Mitaka cycle! 4.229 lines of code refactored and 18.678 lines removed!
• johnthetubaguy: Nova subteams starting to look after themselves, and effectively reporting their progress to the wider community
• All: https://wiki.openstack.org/wiki/Successes

Dropping KEYSTONE_CATALOG_BACKEND – Plus Update Your Devstack Plugins

• Devstack has some half baked support for Keystone templated service catalog.
• In an effort to clean up parts of Devstack, we’re dropping that [2].
• This breaks everyone’s Devstack plugin that references the KEYSTONE_CATALOG_BACKEND variable.
• This variable will be kept around until Newton development opens up.
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-February/086272.html

All Hail the New Per-region PyPI, wheel and APT Mirrors

• We have new AFS-based consistent per-region mirrors of PyPI and API repositories, with additional wheel repositories containing pre-built wheels for all the modules in global-requirements.
• You should not notice a difference, except jobs should be a bit faster and more reliable.
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-February/086295.html

Release countdown for week R-7, Feb 15-19

• Focus: Project teams should be focusing on wrapping up new feature work in all libraries.
• Release Actions:
  • We will be strictly enforcing the library release freeze before Mitaka-3 in 2 weeks.
  • Review client libraries, integration libraries, and any other libraries managed by your team, and ensure recent changes have been released.
  • Ensure global-requirements and constraints lists are up to date, with accurate minimum versions, and exclusions.
  • Projects using cycle-with-intermediary release model need to produce intermediate releases. See Thierry’s emails for details [3].
  • Review stable/liberty branches and submit patches to openstack/releases if you want them.
• Important Dates
  • Final release for  non-client libraries: Feb 24
  • Final release for client libraries: Mar 2
  • Mitaka 3: Feb 29-Mar 4 (includes feature freeze and soft string freeze)
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-February/086362.html

Why WADL When You Can Swagger

• Continuing from previous update [4].
• Every build of the api-site is now running fairy-slipper to migrate from WADL to Swagger.
  • Those migrated Swagger files are copied to [5].
• Not all files migrate smoothly. We’d love to get teams looking at these migrated files. Thank you to those who have already submitted fixes!
  • If you see a problem in the original WADL when viewing [5], log it [7].
  • If you see a problem with the migration tool, log it [8].
• The Infra team is reviewing a specification [6] so that we can serve API information from developer.openstack.org.
• You can hop onto #openstack-doc or #openstack-sdks to ask nick annegentle
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-February/086467.html

Tenant VS. Project

• Sean Dague brings up that OpenStack’s use of tenant vs. project. Which are we transitioning to?
• Keystone is working towards allowing project_id in the service catalog [9].
• Neutron is transitioning to project_id now.
• The current Ansible OpenStack modules are using project [10].
• OSLO Logging/Context are using project.
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-February/086396.html

Proposal: Separate Design Summits From OpenStack Conferences

• The OpenStack design summits originally started out as working events.
• The OpenStack summits growing more marketing and sales focus, the contributors attending are often unfocused.
• Some contributors submit talks for the conference, because their company says it’s the only way for them to attend the conference. Part of the reason for this is the cost of attending.
• Thierry Carrez (who helps organize the design summit) explains that he has been working a solution for separation of the summit and the conference himself, and the Foundation is finalizing a strawman proposal that will be pushed to the community for comments soon.
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-February/086007.html

[1] – http://releases.openstack.org
[2] – https://review.openstack.org/#/c/278333
[3] – http://lists.openstack.org/pipermail/openstack-dev/2016-February/086152.html
[4] – http://www.openstack.org/blog/2016/01/openstack-developer-mailing-list-digest-january-9-15/
[5] – http://developer.openstack.org/draft/swagger
[6] – https://review.openstack.org/#/c/276484/
[7] – https://bugs.launchpad.net/openstack-api-site
[8] – https://bugs.launchpad.net/openstack-doc-tools
[9] – https://review.openstack.org/#/c/279576/
[10] – http://docs.ansible.com/ansible/list_of_cloud_modules.html#openstack

SuccessBot Says

• odyssey4me: OpenStack Ansible Liberty 12.0.5 released.
• stevemar: Devstack now defaults to v3 for Keystone.
• boris-42: osprofiler functional job passed [1].
• odyssey4me: OpenStack Ansible Kilo 11.2.9 released [2].
• odyssey4me: OpenStack Ansible Liberty 12.0.6 released [3].
• All: https://wiki.openstack.org/wiki/Successes

Cross-Project Specs

• A common policy scenario across all projects [4].
• Query config from web UI [5]

API Guidelines

• Must not return service-side tracebacks [6].

Service Type vs. Project Name For Use In Headers

• There’s a question of whether we should be using service type or project names in headers. Some reviews involving this [7][8][9][10].
• We should be selecting things that better serve the API consumers and according to Dean Troyer the API working group is going in the right direction.
• The service type as the primary identifier for endpoints and API services is well established, and is how the service catalog has and will always work. Service types therefore should be the way to go.
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-January/085145.html

OpenStack Ansible Without Containers

• Gyorgy annouces a new installer for OpenStack under GPLv3 using Ansible, but without containers.
  • Reasons for another installer since we already have the OpenStack Ansible project and Kolla:
    • Containers adding unnecessary complexity.
    • Packages: avoid mixing pip and distributor packages. Distributor packages includes things like init scripts, proper system users, upgrade possibilities, etc.
      • Kevin Carter mentions that these benefits are actually also included with the OpenStack Ansible project.
• Without containers, upgrading a single controller can be tricky and disruptive since you have to upgrade every service at the same time. Rollbacks are also easier.
• OpenStack Ansible project can already today do deployment without containers using the is_metal=true variable.
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-January/084963.html

Release Countdown for Week R-8, Feb 8-12

• Focus:
  • 2 more weeks before final releases for non-client libraries for this cycle.
  • 3 more weeks before the final releases for client libraries.
  • Projects should focus on wrapping up feature work in all libraries.
• Release Actions:
  • The release team will be strictly enforcing library release freeze before M3 in 3 weeks.
• Important Dates:
  • Final release for  non-client libraries: Feb 24
  • Final release for client libraries: Mar 2
  • Mitaka 3: Feb 29-Mar 4 (includes feature freeze and soft string freeze)
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-February/085705.html

“No Open Core” in 2016

• Before OpenStack had a name, the “four opens” principles were created to define how we operate as a community.
• In 2010 when OpenStack started, it was different from other open source cloud platforms (Eucalyptus) which followed open core strategy of producing a crippled community edition and an “enterprise version”.
• Today we have a non-profit independent foundation, which cannot do an “enterprise edition”.
  • Today member companies build “enterprise products” on top of the apache licensed upstream project. Some are drivers that expose functionality in proprietary components.
• What does it mean to “not do open core” in 2016? What is acceptable and what’s not?
• Thierry Carrez believes it’s time to refresh this of what is an acceptable official project in OpenStack.
  • It should have a fully-functional production grade open source implementation
  • If you need proprietary software of commercial entity to fully use the project, then it should not be accepted in OpenStack as an official project.
    • These projects can still be non-official projects and still be hosted by OpenStack infrastructure.
• Doug Hellmann brings up Poppy [11] which is applying to be an official OpenStack project.
  • A wrapper to content delivery networks, but there is no open source solution.
  • Is this something that can be an official project, or is open core?
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-February/085855.html

The Trouble with Names

• A few issues have crept up recently with the service catalog, API headers, API end points, and even similarly named resources in different resources (e.g. backup), that are all circling around a key problem. Distributed teams and naming collision.
• Each project has a unique name that is ensured by their git repository in the OpenStack namespace.
• There’s a desire to replace project names with generic names like nova/compute in:
  • service catalog
  • api headers
• Options we have are:
  • Use the code names we already have: nova, glance, swift, etc.
    • Upside: collision problem solved.
    • Downside: You need a secret decoder ring to know what a project does.
  • Have a registry of common names.
    • Upside: we can safely use common names everywhere and not fear collision down the road.
    • Downside: yet another contention point.
• Approvals by the various people in the community to have a registry of the common names. Maybe in the governance projects.yaml file [12]?
  • This list does include only the official projects by the technical committee, therefore only those projects can reserve the common names.
• OpenStack Client has already encoded some of these common names to projects [13].
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-February/085748.html

Announcing Ekko – Scalable Block-Based Backup for OpenStack

• The goal of Ekko is to provide incremental block-level backup and restore of Nova instances.
• Two place with overlapping goals:
  • Cinder volume without having the incremental backups be dependent.
  • Nova instances
    • OpenStack Freezer today leverages Nova’s snapshot feature.
    • Ekko would leverage live incremental block-level backup of a nova instance.
• Jay Pipes begins the discussion on the two projects (Freezer and Ekko) working together to make sure their REST API endpoints are not overlapping. Having two APIs for performing backups that are virtually identical is not good.
• The creator of Ekko sees the reason for another backup project because of “actual implementation and end results are wildly different” even if they are the same API call.
• Jay doesn’t like that today all the following endpoints exist:
  • Freezer’s /backups
  • Cinder’s /{tenant_id}/backups
• Having these endpoints make for bad user experience and is just confusing.
• The current governance model does not prevent competition of projects. So if two projects overlap in API endpoints, there should be an attempt in collaboration.
• Full thread: http://lists.openstack.org/pipermail/openstack-dev/2016-January/084739.html