Loading…
This event has ended. Create your own event → Check it out
This event has ended. Create your own
This is the schedule for the Kilo Design Summit, where OpenStack contributors discuss the future of OpenStack development.
Click here for the main OpenStack Summit conference schedule.
View analytic

Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

Monday, November 3
 

11:40

Ops Summit: Top 10 Pain points from the user survey - how to fix them?
Etherpad: https://etherpad.openstack.org/p/kilo-summit-ops-pain

Every survey, the user committee compiles a list of the top pain points in using OpenStack. Come and see what they are, and let's work together to make a plan to get them fixed to they don't come up again next time!

Don't agree with the pain points listed? Add you own on the etherpad link!

Moderated by the OpenStack User Committee.

Moderators
avatar for Subbu Allamaraju

Subbu Allamaraju

Chief Engineer, Cloud, eBay Inc
Subbu is the Chief Engineer of cloud at eBay Inc. His team builds and operates a multi-tenant geographically distributed OpenStack based private cloud. This cloud now serves 100% of PayPal web and mid tier workloads, significant parts of eBay front end and services, and thousands of users for their dev/test activities.
avatar for Tim Bell

Tim Bell

Infrastructure Services Manager, CERN
Tim is currently responsible for the team at CERN that manages the operating system and infrastructure services. He previously worked as a Unix kernel developer at IBM and managing large-scale Unix production deployments and services for Deutsche Bank. As part of CERN's data centre expansion, Tim has been involved in the implementation of the CERN private cloud and is an elected member of the OpenStack management board.

Speakers
avatar for Jonathan Proulx

Jonathan Proulx

Sr Technical Architect - Computer Science & Artificial Intelligence Lab, MIT


Monday November 3, 2014 11:40 - 12:20
Manet

11:40

Ops Summit: What is best practice for managing 'pets'? (single instances with high uptime requirements)
Etherpad: https://etherpad.openstack.org/p/kilo-summit-ops-pets

Do you have a need to run 'pets' on OpenStack? How do you do it right now? How would it work in your ideal world?

Moderators
avatar for Russell Bryant

Russell Bryant

Senior Principal Software Engineer, Red Hat
Russell Bryant is Technical Director for OpenStack at Red Hat.  Russell has been working on OpenStack since late 2011. He has primarily worked in the areas of Compute and Networking. Russell served as the Project Technical Lead for Nova (Compute) for two releases. He also helped build the Open Virtual Network (OVN) project and is an Open vSwitch project committer. He has also been elected to the OpenStack Technical Committee and Foundation... Read More →
avatar for Adam Lawson

Adam Lawson

Cloud Solutions Architect, AQORN INC
Nothing about cloud architecture and big data that I don't love. If you find me at one of the Summits, I'm usually designing for a large company or busy working on a new feature in the Thunder openstack deployment platform (http://www.aqorn.com/products) so don't be shy; I love to meet new people! | | //adam

Monday November 3, 2014 11:40 - 12:20
Degas

11:40

Ceph for OpenStack Design Session
Ceph is one of the most popular storage option for OpenStack installations. We will use this time and space to discuss future Ceph enhancements related to the OpenStack Kilo release.

Monday November 3, 2014 11:40 - 13:10
Dufy Le Meridien

11:40

OpenDaylight Neutron Plugin Design Session
OpenDaylight is the leading open source platform for Software Defined Networking (SDN). Currently there is an ML2 plugin for OpenDaylight within Neutron. Significant work has already been done to show the benefits of managing virtual and physical network elements via OpenStack through the OpenDaylight controller. In addition, service function chaining, group-based-policy, security-groups and distributed virtual routing are all development activities that span both OpenStack and OpenDaylight and deeper collaboration on these topics will significantly benefit both projects.

The bulk of the time will be used in discussion of the above activities and brainstorming plans for work in Kilo and Lithium (OpenDaylight’s next release) to improve the network programmability available to OpenStack through the use of OpenDaylight. Additional technical topics for discussion will also be solicited from members of the audience and pursued as time permits.

Monday November 3, 2014 11:40 - 13:10
Palais - Room 124/125 Palais des Congrès

12:30

Ops Summit: Distributed Virtual Router - finally neutron HA!
A presentation for ops folks about what Neutron DVR actually does!

Etherpad: https://etherpad.openstack.org/p/kilo-summit-ops-dvr

Moderators
EM

Edgar Magana

Cloud Operations Architect, Workday
Edgar is an emerging leader who has specialized in Cloud Computing, Network Virtualization, Software-defined Networking (SDN), Network Functions Virtualization (NFV), OpenFlow and OpenStack. He has developed excellent software development skills and outstanding customer and business driven experience. Currently, he is core member of the Neutron OpenStack development community. Edgar has strong experience in fully automated Cloud Computing... Read More →

Monday November 3, 2014 12:30 - 13:10
Degas

12:30

Ops Summit: How do we fix logging? Bring your examples of bad!
Isn't logging in OpenStack annoying? If you don't have debug on, you can't get what you need. Even with debug off, there's too many irrelevant messages. The format isn't consistent between services, and logs are diffcult to trace.

So add your pet hates on the etherpad and let's fix this!

Etherpad: https://etherpad.openstack.org/p/kilo-summit-ops-logging

Moderators
avatar for Rochelle Grober

Rochelle Grober

Senior Staff Architect - Huawei, Huawei Technologies
Rocky is an industry veteran, with experience spanning computer bring up to AI, networks and embedded. But her attention always seems to return to close to the metal, large infrastructure. Starting out in EE, she migrated to SW development then on to QA and SW Process, which is why she is a champion for operations, interoperability and usability. She has been aware and peripherally involved in Open Source since the original creation of Copy Left... Read More →
avatar for Jay Pipes

Jay Pipes

Director of Upstream Engineering, Mirantis
Jay is Director of Upstream Engineering at Mirantis. He manages a team of senior engineers working on developing OpenStack projects in the community. Before Mirantis, he worked at AT&T, focusing on OpenStack development and operations. Previously, he was a Director of Engineering in HP s Open Source Cloud Services team, a development manager and software engineer at Rackspace Cloud, a software engineer at Sun Microsystems and the North American... Read More →

Monday November 3, 2014 12:30 - 13:10
Manet

13:10

Lunch
Monday November 3, 2014 13:10 - 14:30
Le Bar Le Meridien

14:30

Ops Summit: What do you want from ironic/bare metal?
Ironic developers will attend to lead a discussion around the bare metal component in OpenStack. Bring your bugs and your wishlist!

https://etherpad.openstack.org/p/kilo-summit-ops-ironic

Moderators
avatar for Devananda Van Der Veen

Devananda Van Der Veen

Bare Metal Cloud Architect, IBM Cloud / SoftLayer
Devananda is opinionated and passionate about using technology to improve humanity. He began working on OpenStack in 2012 and started the Ironic project a year later, adding bare metal provisioning to the growing cloud platform, and subsequently served on the OpenStack Technical Committee. | | When not working from home in the cloudy city of Seattle, he can be found speaking at open source / tech conferences.

Monday November 3, 2014 14:30 - 15:10
Manet

14:30

The Application Ecosystem Working Group
https://wiki.openstack.org/wiki/Application_Ecosystem_Working_Group

* Welcome
* Where do you hang out? How can we reach you?
* What do you do? Anyone able to give a lightening intro?
* Feedback on developer.openstack.org and the type of content that would be useful
* User survey - "Your OpenStack Usage" section - results and tweaks
* SDKs & working with SDK communities
* Where are the bugs/rough edges in OpenStack for you?
* Getting feedback into the development process

Etherpad: https://etherpad.openstack.org/p/kilo-summit-ops-app-eco

Moderators
avatar for Tim Bell

Tim Bell

Infrastructure Services Manager, CERN
Tim is currently responsible for the team at CERN that manages the operating system and infrastructure services. He previously worked as a Unix kernel developer at IBM and managing large-scale Unix production deployments and services for Deutsche Bank. As part of CERN's data centre expansion, Tim has been involved in the implementation of the CERN private cloud and is an elected member of the OpenStack management board.
avatar for Tom Fifield

Tom Fifield

OpenStack community manager
After working on scalability in computing at particle physics experiments like ATLAS at the Large Hadron Collider, Tom led the creation of a cloud designed for the publicly-funded research sector in Australia.It currently serves thousands of researchers directly, using many datacentres over a broad geographical area. | Following working as a cloud architecture consultant and team lead for several years, Tom is currently harnessing his passion... Read More →
avatar for Chris C Kemp

Chris C Kemp

Founder of Nebula & OpenStack, Nebula
Chris founded and launched Nebula in July of 2011 to disrupt enterprise computing and ignite a new era of global innovation after serving as NASA's Chief Technology Officer. At NASA, Chris co-founded the OpenStack project and led the Nebula project, NASA's cloud computing initiative. Chris also served as the Chairman of the White House Federal Cloud Computing Standards Working Group and was a member of the Cloud Computing Executive Steering... Read More →
avatar for Everett Toews

Everett Toews

Developer Advocate, Rackspace
Everett Toews is a Docker Captain and a Developer Advocate at Rackspace making the cloud easy to use for developers and operators. Sometimes dev, sometimes ops, and sometimes advocate. He's interested in composing systems using APIs, SDKs, CLIs, and Docker. He likes to teach what he learns along the way.

Monday November 3, 2014 14:30 - 15:10
Degas

14:30

Ansible for OpenStack Design Session
Ansible is an IT automation tool. It can configure systems, deploy software, and orchestrate more advanced IT tasks such as continuous deployments or zero downtime rolling updates.

Ansible’s goals are foremost those of simplicity and maximum ease of use. It also has a strong focus on security and reliability, featuring a minimum of moving parts, usage of OpenSSH for transport (with an accelerated socket mode and pull modes as alternatives), and a language that is designed around auditability by humans – even those not familiar with the program.

Ansible is a popular tool for deploying and managing OpenStack, as well as for deploying and managing workloads run on OpenStack.

There are three areas of interest in the intersection of ansible and OpenStack.

- deploying OpenStack with ansible
- OpenStack support in ansible/managing workloads on OpenStack with ansible
- use of ansible in OpenStack itself

All three are worthy topics, but the second two are potentially more useful to discuss in person. Namely, what, if anything, does the OpenStack community want to do in terms of ensuring a good enduser experience for ansible users on OpenStack, and where, if anywhere, can ansible be useful inside of OpenStack.

Monday November 3, 2014 14:30 - 16:00
Dufy Le Meridien

14:30

Cloud Foundry BOSH CPI for OpenStack
Cloud Foundry BOSH Cloud Provider Interface or CPI provides implementation and is designed to manage the deployment of Cloud Foundry on top of OpenStack.

In this session we will discuss what other enhancements could be made to the CPI to automate and further enhance Cloud Foundry deployments on OpenStack.

Moderators
avatar for Ferran Rodenas

Ferran Rodenas

Cloud Foundry Platform Engineer, Pivotal
avatar for Animesh Singh

Animesh Singh

STSM and Lead Architect, IBM
Animesh Singh is an STSM and Lead for IBM Cloud Platform and Infrastructure. He has been with IBM for more than 10 years and currently works with customers in designing cloud computing solutions on OpenStack, Cloud Foundry, Docker and Serverless Architecture, and is the lead for IBM Bluemix. He has been leading cutting edge projects for IBM enterprise customers in Telco, Banking, and Healthcare Industries, around cloud and virtualization... Read More →

Monday November 3, 2014 14:30 - 16:00
Palais - Room 124/125 Palais des Congrès

15:20

Ops Summit: Architecture Show and Tell - Upgrades Special Edition
Come and see some lightening talks from people who've gone through the pain of doing an OpenStack Upgrade!

https://etherpad.openstack.org/p/kilo-summit-ops-architecture-upgrades

Speakers
avatar for Subbu Allamaraju

Subbu Allamaraju

Chief Engineer, Cloud, eBay Inc
Subbu is the Chief Engineer of cloud at eBay Inc. His team builds and operates a multi-tenant geographically distributed OpenStack based private cloud. This cloud now serves 100% of PayPal web and mid tier workloads, significant parts of eBay front end and services, and thousands of users for their dev/test activities.
avatar for Tim Bell

Tim Bell

Infrastructure Services Manager, CERN
Tim is currently responsible for the team at CERN that manages the operating system and infrastructure services. He previously worked as a Unix kernel developer at IBM and managing large-scale Unix production deployments and services for Deutsche Bank. As part of CERN's data centre expansion, Tim has been involved in the implementation of the CERN private cloud and is an elected member of the OpenStack management board.


Monday November 3, 2014 15:20 - 16:00
Degas

15:20

Ops Summit: What do we want from containers/docker?
So far, we have the nova docker driver in stackforge, but it's not enough. Let's look at what we need for containers in OpenStack.
https://etherpad.openstack.org/p/kilo-summit-ops-containers

Moderators
MS

Michael Still

Software Engineer, Rackspace
Michael Still is a Nova and Oslo core reviewer at Rackspace, where he works on the Open Source OpenStack project as part of the Private Cloud team. He spends most of his time hacking on the libvirt virtualization layer in nova. | Before joining Rackspace in 2012, Michael spent six years as a Site Reliability Engineer at Google and one year as an Operations Engineer at Canonical. In both roles, he was responsible for maintaining and improving web... Read More →

Monday November 3, 2014 15:20 - 16:00
Manet

16:00

Break
Monday November 3, 2014 16:00 - 16:20
Developer lounge Le Meridien

16:20

Ops Summit: Architecture Show and Tell
More Lightening talks, showing off interesting architecture aspects, as well as tales of fail!

https://etherpad.openstack.org/p/kilo-summit-ops-architecture

Moderators
avatar for Tom Fifield

Tom Fifield

OpenStack community manager
After working on scalability in computing at particle physics experiments like ATLAS at the Large Hadron Collider, Tom led the creation of a cloud designed for the publicly-funded research sector in Australia.It currently serves thousands of researchers directly, using many datacentres over a broad geographical area. | Following working as a cloud architecture consultant and team lead for several years, Tom is currently harnessing his passion... Read More →

Speakers
avatar for Mike Dorman

Mike Dorman

Senior Systems Engineer, Go Daddy
Mike is a senior systems engineer at Go Daddy, building and managing OpenStack cloud infrastructure and supporting systems.  He thrives on enabling others to work faster and better by using the cloud.  Mike lives in Colorado with his wife and three children, and enjoys hiking (summer) and snowshoeing (winter.)  Follow Mike on Twitter @misterdorm, or on his blog, http://dorm.org
avatar for Wade Lewis

Wade Lewis

OpenStack Dedicated Architect
Wade Lewis is an Openstack Dedicated Architect for Rackspace Private Cloud. He's been a Racker for 12 years and has been involved with large scale Openstack clouds since Folsom. His current projects involve working with financial and medical institutions that are migrating legacy workloads to the cloud. 
avatar for Kris Lindgren

Kris Lindgren

Senior Linux Systems Engineer, Godaddy
Kris Lindgren is an operator of Go Daddy's public and private cloud.  He has been using openstack in a production capactiy since Havana.  He loves enabling others to work faster and better by using the cloud and other self-service portals.  He is also a core dev on the anvil project ( https://github.com/stackforge/anvil ), which allows deployers to package upstream openstack + their local patchsets into system level packages or virtual... Read More →
avatar for Jonathan Proulx

Jonathan Proulx

Sr Technical Architect - Computer Science & Artificial Intelligence Lab, MIT
avatar for Jason Smith

Jason Smith

Sr. Linux Systems Engineer (Enterprise Segment), Rackspace
Jason has been in the IT industry for over 10 years.  He has often overcome challenging problems with innovative, creative solutions. Faced with difficult challenges such as technology adoption, integration of high performance clusters, and systems deployment automation, Jason has led the pack on implementing new technologies and teaching best practices. | | Jason has spent the better part of his career managing dedicated servers through... Read More →
avatar for Joe Topjian

Joe Topjian

Systems Architect, Cybera
Joe currently lives in Alberta, Canada, building and automating clouds for Cybera. Joe is also a co-author of the OpenStack Operations Guide and is an active member of the OpenStack Operators community.


Monday November 3, 2014 16:20 - 17:00
Manet

16:20

Ops Summit: High Availability - how do you do it?

Moderators
avatar for Andrew Mitry

Andrew Mitry

Distinguished Engineer, Walmart
Andrew Mitry is currently a cloud lead at Walmart. | | Andrew Mitry lead Comcast’s OpenStack team for the past three years where both the infrastructure and team have grown more than 500% in the past year. The team led Internet Protocol version 6 (IPv6) in OpenStack and ranked as a top 20 contributor with more than 36,000 lines of code.

Monday November 3, 2014 16:20 - 17:00
Degas

16:20

Puppet-OpenStack Design Session
The puppet-openstack modules on StackForge are an effort to address both the deployment and long-term configuration and maintenance of an OpenStack system.

During this session, we will create a roadmap for the next six-month release cycle, addressing issues including:

* Automated integration testing
* Features of the openstacklib and openstack_extras modules, such as HA support and other common library resources
* Review of core members
* Composition module
* Release management and namespace
* Identifying valuable platform integrations beyond Puppet modules
* Any other topics proposed by the community.

Monday November 3, 2014 16:20 - 17:50
Palais - Room 124/125 Palais des Congrès

16:20

Solum for OpenStack Design Session
Solum allows cloud operators to add customizable CI/CD Pipeline capability to applications so they can be built, tested, and run on OpenStack clouds. This design session will start with a Solum demo, then move on to planning feature development for the OpenStack Kilo release.

Monday November 3, 2014 16:20 - 17:50
Dufy Le Meridien

17:10

Ops Summit: How to get involved in Kilo
As an OpenStack user, you have a serious amount of influence over the direction of the development roadmap. Come along and see how you can get involved, and hear from some prominent people in the development community on where they need your cooperation.

https://etherpad.openstack.org/p/kilo-summit-ops-get-involved

Moderators
avatar for Tom Fifield

Tom Fifield

OpenStack community manager
After working on scalability in computing at particle physics experiments like ATLAS at the Large Hadron Collider, Tom led the creation of a cloud designed for the publicly-funded research sector in Australia.It currently serves thousands of researchers directly, using many datacentres over a broad geographical area. | Following working as a cloud architecture consultant and team lead for several years, Tom is currently harnessing his passion... Read More →
avatar for Stefano Maffulli

Stefano Maffulli

Community Manager, OpenStack Foundation, OpenStack Foundation
| OpenStack community manager, Stefano built his career around Free Software and open source: from pre-sales engineer and product manager at Italian GNU/Linux distribution MadeInLinux to Italian Chancellor of the Free Software Foundation Europe, where he also created the FSFE Fellowship participation program. Later as community manager of leading mobile open source sync solution Funambol, his efforts boosted downloads and bolstered enterprise... Read More →

Monday November 3, 2014 17:10 - 17:50
Manet

17:10

Ops Summit: Storage - how do you do it?
Let's share storage best practices, and bug reports.
https://etherpad.openstack.org/p/kilo-summit-ops-storage

Moderators
avatar for Warren Wang

Warren Wang

Cloud Engineer, Walmart
Warren is a cloud engineer at Walmart. He has focused on OpenStack and Ceph for the last few years. Warren enjoys working on large scale projects, like he did at AOL and Comcast, and continues to do so at Walmart.

Monday November 3, 2014 17:10 - 17:50
Degas
 
Tuesday, November 4
 

11:15

API Working Group (Part 1)
The API Working Group's purpose is to propose, discuss, review, and advocate for API guidelines for all OpenStack Programs to follow. In this session we'll be discussing the API Working Group ( https://wiki.openstack.org/wiki/API_Working_Group ) itself, including reviewing sections on the wiki page, our processes, and how we can be an effective working group.

Etherpad: https://etherpad.openstack.org/p/kilo-crossproject-api-wg

Session Lead: Jay Pipes, Everett Toews


Tuesday November 4, 2014 11:15 - 11:55
Manet

11:15

DefCore, RefStack, Interoperability, and Tempest
Session Leads: David lenwell, Catherine Diep, Rocky Grober


Tuesday November 4, 2014 11:15 - 11:55
Degas

11:15

Incremental Zone Transfers with designate-mdns
Incremental Zone Transfers with mDNS


Tuesday November 4, 2014 11:15 - 11:55
Gauguin

11:15

Design Summit 101

Is this your first OpenStack Design Summit? Unlike most conferences, you are invited to participate and play an active role. But... where to start? The rationale and the organization that allows such a unique collaborative design will be explained. This is your chance to get answers and get the best of it!

During this session we will let the attendants know some details about the Design Summit, including who will be attending, different tracks and purposes, which sessions/talks are the most suitable for beginners and how they can participate. This short introduction will be followed by a a lively presentation of the most common situations and how to behave when facing them. It will be a miniature experience of a first participation to the Design Summit.


Tuesday November 4, 2014 11:15 - 11:55
Derain

11:15

Rally / OSProfiler
Tuesday November 4, 2014 11:15 - 11:55
Dufy Le Meridien

11:15

Zaqar integration with other services
https://etherpad.openstack.org/p/kilo-zaqar-summit-integration-with-services


Tuesday November 4, 2014 11:15 - 11:55
Corot

11:15

OpenStack-Chef Design Session
Presentation of how Chef can be used to manage OpenStack deployments. Topics include:

* The future roadmap
* Community Concerns and Questions
* Q & A 

Tuesday November 4, 2014 11:15 - 12:45
Palais - Room 124/125 Palais des Congrès

12:05

API Working Group (Part 2)
The API Working Group's purpose is to propose, discuss, review, and advocate for API guidelines for all OpenStack Programs to follow. In this session we'll discuss pain points and inconsistencies in the OpenStack APIs and propose guidelines to improve those problems. To be concluded on Thursday in http://kilodesignsummit.sched.org/event/3f0a5f22f2d641ef69965373f3e23983

Etherpad: https://etherpad.openstack.org/p/kilo-crossproject-api-wg

Session Lead: Jay Pipes, Everett Toews


Tuesday November 4, 2014 12:05 - 12:45
Manet

12:05

Common Approach to HA
Let's try and get some agreement on what projects should do (and hopefully backed by tests) to make services behave correctly from a cluster manager's perspective.

Etherpad: https://etherpad.openstack.org/p/kilo-crossproject-ha-integration

Session Lead: Angus Salkeld


Tuesday November 4, 2014 12:05 - 12:45
Derain

12:05

Moving Functional Tests to Projects
Our current methodology of using tempest to house functional tests which are run against all projects on every commit has several problems
* It puts the testing focus for all projects on the tempest team which is not scalable
* There is not time to run deep functional tests because too much time is spent running tests against projects that cannot be affected
The move to per-project functional testing, combined with more aggressive post-merge testing, will make catching bugs per-commit more likely
and reduce the chance that bad code slipping in to some project will impact the gate stability for other projects. Some issues that need resolution:
* How do we retain the ability to export a functional test suite for OpenStack as a whole, the way tempest is used now? This is important for
refstack and any one else doing validation.
* What kinds of non-scenario, non-integration tests remain in tempest?
** None?
** Some coverage for API stability?
* How do we make sure projects do their testing in a consistent way? Do they need to?
* Where do two-project integration tests live?

Session Lead: David Kranz


Tuesday November 4, 2014 12:05 - 12:45
Degas

12:05

DNSSEC with Designate
DNSSEC with Designate


Tuesday November 4, 2014 12:05 - 12:45
Gauguin

12:05

Group-Based-Policy
Group Based Policy (GBP) is a StackForge project that provides declarative abstractions for achieving scalable intent-based infrastructure automation. GBP was originally conceived in the Atlanta Design Summit and has since evolved to its current state where it can be put into the hands of early adopters.

Currently GBP complements the OpenStack networking model with the notion of policies that can be applied between groups of network endpoints. As users look beyond basic connectivity, richer network services with diverse implementations and network properties are naturally expressed as policies. Examples include service chaining, QoS, path properties, access control, etc. GBP thus allows application administrators to express their networking requirements using a Group and a Policy Rules-Set abstraction. The specifics of policy rendering are left to the underlying pluggable policy driver. The GBP model also inherently supports constructs that makes it easy to abstract and consume complex network service chains and graphs.

This model of defining groups of entities and providing abstractions to capture user intent in the form of a policy language is being extended to the entire infrastructure in general. In this session we will review the GBP implementation for Juno, discuss the roadmap for Kilo, and also share pointers on how new participants can contribute to the project.

Wiki page: https://wiki.openstack.org/wiki/GroupBasedPolicy

Tuesday November 4, 2014 12:05 - 12:45
Dufy Le Meridien

12:05

Zaqar API v2: What? Why? When?
https://etherpad.openstack.org/p/kilo-zaqar-summit-v2


Tuesday November 4, 2014 12:05 - 12:45
Corot

12:45

Lunch
Tuesday November 4, 2014 12:45 - 14:00
Le Bar Le Meridien

14:00

Consistent Approach to Upgrades and Versioning
Etherpad: https://etherpad.openstack.org/p/kilo-crossproject-upgrades-and-versioning

Let's explore the soon to be oslo-versioned-objects and what it can (and can't) provide a project in terms of supporting upgrades.

Session Leads: Angus Salkeld, Dan Smith


Tuesday November 4, 2014 14:00 - 14:40
Derain

14:00

Schema and Schema Validation for Notifications
Notifications emitted from OpenStack Services have a small number of required fields: event_type, timestamp and priority. The rest goes in an undefined "payload" section. As OpenStack has grown we've seen the need to enforce the schema of these payload sections as well as establish some consistency rules for notifications. This session will define a plan for implementing this schema validation as well as how to deprecate the old notification scheme (or provide an alternative). The challenge to migration is updating all the code in all the OpenStack services which hand-crafts the payload data. What will happen to existing downstream systems that expect the current format?

Session Leads: Sandy Walsh, Julien Danjou


Tuesday November 4, 2014 14:00 - 14:40
Degas

14:00

Specs: The Good, Bad, and the Ugly
Most projects have adopted a new process where we use gerrit to review design specifications associated with blueprints. After using this process for the Juno release, we should get back together and discuss what is going well or not so well. We can discuss potential changes and opportunities for standardizing the approach across projects. Finally, we can also discuss the addition of a cross-project specs repo for the Kilo release.

Session Lead: John Garbutt


Tuesday November 4, 2014 14:00 - 14:40
Manet

14:00

Designate: Standalone Agent
Replacement Standalone Agent for the mDNS world


Tuesday November 4, 2014 14:00 - 14:40
Gauguin

14:00

Poppy
Poppy: An Open API for CDN Provisioning

OpenStack operators have many choices when incorporating a Content Delivery Network (CDN) into their infrastructure -- the CDN marketplace has both tried-and-true vendors and up-and-coming upstarts with innovative new features.

But these vendors often have highly-customized and proprietary provisioning APIs. This can be problematic when an operator wishes to support multiple providers -- or swap out one vendor for another. And these challenges spill over to developers who become forced into codifying the CDN instructions for multiple vendors into their applications.

Poppy aims to solve these challenges. Written as a modular, vendor-neutral API, Poppy incorporates a driver-based model that wraps provisioning instructions for all CDN vendors that support it. Application developers can write their code once, and Poppy will handle all the requisite translations behind-the-scenes.

Poppy's success depends on active and engaged participation by CDN Providers. Much like Cinder, Manila, and Designate use a driver-based approach to connect multiple vendors to a consistent API, Poppy will do the same for CDN Providers.

This design session will discuss with the openstack community and CDN providers on how to coalesce on the ideal vendor-neutral API, and lay the implementation plans for the platform and driver development in preparation for OpenStack Kilo.

As you'll see in the links below, multiple CDN providers are already engaged, and significant design, documentation, and implementation work is underway.

http://poppycdn.org/
https://wiki.openstack.org/wiki/Poppy
https://github.com/stackforge/poppy

Tuesday November 4, 2014 14:00 - 14:40
Dufy Le Meridien

14:00

Zaqar Persistent Transports
https://etherpad.openstack.org/p/kilo-zaqar-summit-persistent-transports


Tuesday November 4, 2014 14:00 - 14:40
Corot

14:00

MagnetoDB, OpenStack Key-Value Store, Design Session
MagnetoDB is a key-value storage service, designed and developed as an OpenStack service. It provides a similar storage as a service model to OpenStack Swift or Cinder, providing a data path ReST API for easy data storage and retrieval. It uses OpenStack Keystone for authorization and enforces secure multi-tenancy using OpenStack's model of domains and projects. MagnetoDB leverages OpenStack Notifications for integration with billing and monitoring systems.

The overview of project achievements during Juno cycle will be presented. The roadmap for Kilo will be discussed, and we will be diving into the design and implementation details of major features.

Tuesday November 4, 2014 14:00 - 15:30
Palais - Room 124/125 Palais des Congrès

14:50

Integrating Barbican with other OpenStack projects
This is a discussion for strategies and patterns to be used when integrating your OpenStack project with Barbican. The Barbican team has seen a couple of different approaches to integration in Nova, Glance, and Neutron LBaaS. This discussion aims to find a consistent path for integration that interested projects can use to minimize code repetition.


Tuesday November 4, 2014 14:50 - 15:30
Gauguin

14:50

Approaches for Scaling Out
Nova has had support for Cells for several releases now. There is another approach, Cascading, that has been recently proposed. Let's quickly recap cells and then discuss the cascading alternative.

Session Leads: John Garbutt, Chaoyi Huang


Tuesday November 4, 2014 14:50 - 15:30
Manet

14:50

Changes to our Requirements Management Policy
As we move to testing lots of projects under a devstack configuration, and possibly reduce the integrated release scope, we need to support requirements from projects that are not integrated. These projects need to use a common set of dependencies to allow them to be installed together for deployment, even if they are not tested together in our gate.

Session Lead: Doug Hellmann


Tuesday November 4, 2014 14:50 - 15:30
Derain

14:50

Scaling Documentation Across Projects
To borrow from the mailing list thread at http://lists.openstack.org/pipermail/openstack-dev/2014-October/047801.html "The current system is something of a hybrid model - for some subset of official projects considered "important", the Docs team is directly responsible; for the others, the project team has to write the documentation. The docs team is available to provide support and tools for other official projects." Some questions for discussion:

Project liasons, what should they do? The OpenStack Documentation is centralized on docs.openstack.org but often there's a need for specialty information when reviewing patches or triaging doc bugs. A doc liaison should be available to triage doc bugs when the docs team members don't know enough to triage accurately, and be added to doc reviews that affect your project. You'd be notified through email when you're added either to a doc bug or a doc review. We also would appreciate attendance at the weekly doc team meeting, We meet weekly in #openstack-meeting every Wednesday at alternating times.

Do project teams want a Documentation team as advisory? What expectations do you have from the Docs team?

How should the Docs team engage projects that are not already writing user/operator/API docs? How to get integrating teams better docs faster?

How should we write, review, and publish integrated documentation?

Session Lead: Anne Gentle


Tuesday November 4, 2014 14:50 - 15:30
Degas

14:50

Compass
(Speakers: Shuo Yang, Weidong Shao)

In this session, we will
  • quickly demo (10 mins) of OpenStack deployment with Compass and explain what happened behind the scene.
  • come and play compass with us, and we will stand up an OpenStack on your a Linux/Mac/Windows box - we will have a feature collection and design brainstorming session. Specifically, we would like to share the following design topics
  • meta-data description of a target system (this is to enable existing Cookbooks/Manifests/Playbooks to be plugged into Compass)
  • exploring the hardware management engines (we plan to use existing frameworks such as xCat, Ironic IPA, ONIE)
  • we would like hear your input.

Please provide your interest for discussion in this session at design session Etherpad entry (https://etherpad.openstack.org/p/compass_design_collections )

Tuesday November 4, 2014 14:50 - 15:30
Dufy Le Meridien

14:50

Zaqar Infrastructure Session
https://etherpad.openstack.org/p/kilo-zaqar-summit-infrastructure


Tuesday November 4, 2014 14:50 - 15:30
Corot

15:40

Barbican Common Certificate Issuance API
This discussion aims to define the API that will be used to issue certificates across multiple Certificate Authorities. There are two approaches being discussed: One is to have a baseline API that can be used to provision certificates on any CA, the other approach would introduce a new concept akin to VM "flavors" that can be used to discover which CAs are supported by a particular instace of Barbican.


Tuesday November 4, 2014 15:40 - 16:20
Gauguin

15:40

End user experience, SDKs
In a recent large discussion about development priorities in OpenStack, a major theme was improving end user experience. Let's discuss some ideas of things that could be accomplished in Kilo to improve the situation.

Session Lead: Monty Taylor


Tuesday November 4, 2014 15:40 - 16:20
Derain

15:40

How to Tackle Technical Debt in Kilo
Session Lead: Mark McClain


Tuesday November 4, 2014 15:40 - 16:20
Manet

15:40

Log Rationalization
https://etherpad.openstack.org/p/kilo-crossproject-log-rationalization

Session Lead: Sean Dague


Tuesday November 4, 2014 15:40 - 16:20
Degas

15:40

Manila Networking and Multitenancy
This session is to discuss changes to Manila to support a wider variety of network configurations and multitenancy scenarios.


Tuesday November 4, 2014 15:40 - 16:20
Corot

15:40

Mistral
Tuesday November 4, 2014 15:40 - 16:20
Dufy Le Meridien

16:20

Break
Tuesday November 4, 2014 16:20 - 16:40
Developer lounge Le Meridien

16:40

Barbican Plugin (Driver) Lifetime management
This session aims to define how the various different plugins (drivers) will be managed throughout their lifetime. Some of the questions we aim to answer are: How to migrate from one plugin to another? How to deal with plugin versioning? When does it make sense to enable more than one type of plugin at the same time?


Tuesday November 4, 2014 16:40 - 17:20
Gauguin

16:40

Gerrit third-party CI discussion
This discussion will focus on priorities third party operators have identified as needing further community input (https://etherpad.openstack.org/p/kilo-third-party-items) which have been discussed at the weekly third party meeting (https://wiki.openstack.org/wiki/Meetings/ThirdParty).

Tuesday November 4, 2014 16:40 - 17:20
Derain

16:40

Growth Challenges (Part 1)
After four years of continued expansion, the OpenStack project is encountering new growth challenges. Those call for rapid and aggressive solutions to ensure that we keep on growing and adapt to an ever-changing world in the future. In this cross-project workshop we'll explore various growth challenges (scaling core teams inside projects, scaling the gate, scaling the horizontal teams, how to get competition without duplication...), brainstorm solutions, and expose the current proposed reforms (big tent, layer 1...).



Tuesday November 4, 2014 16:40 - 17:20
Manet

16:40

Keystone Feature Adoption
Keystone has introduced a number of new features over time. Many of these features are never consumed by the other services in the OpenStack ecosystem. For the features that are meant to be utilized by the other services, we need to have a clear path to drive adoption. What are the short term and long term plans for consuming these features? A specific focus on the following:

Etherpad: https://etherpad.openstack.org/p/kilo-crossproject-keystone-feature-adoption

Session Leads: Jamie Lennox, Morgan Fainberg


Tuesday November 4, 2014 16:40 - 17:20

16:40

Manila Mount Automation
The topic of automating mount operations has come up over and over during talks about Manila. There seems to be wide agreement that something is needed, and there are a lot of ideas for how to make this work. Kilo is the right time to solve this problem.


Tuesday November 4, 2014 16:40 - 17:20
Corot

16:40

Murano
Murano Project introduces an application catalog, which allows application developers and cloud administrators to publish various cloud-ready applications in a browsable categorized catalog. It may be used by the cloud users (including the unexperienced ones) to pick-up the needed applications and services and composes the reliable environments out of them in a "push-the-button" manner.

In this session we will briefly overview results of Juno release cycle and focus the discussion on our goals for Kilo release cycle.

Session etherpad: https://etherpad.openstack.org/p/kilo-murano-design-session

Tuesday November 4, 2014 16:40 - 17:20
Dufy Le Meridien

16:40

Congress, Policy Engine for OpenStack, Design Session
Design sessions with Neutron GBP, ODL, Swift, Congress, Nova, and Heat.

Tuesday November 4, 2014 16:40 - 18:10
Palais - Room 124/125 Palais des Congrès

17:30

Translators / I18N team meetup
The Internationalization (I18N) team would like to invite all translators to the team meetup.

Tuesday November 4, 2014 17:30 - 18:10
Derain

17:30

Barbican per-user entity-level authorization
The goal of this session is to find the correct OpenStack-wide way forward for providing entity-level authorization (authorization for individual secrets/containers/etc.) There has been a lot of discussion around this topic since last summit without a resolution. Keystone policy currently cannot address this use case, and there is some resistance to adding it to Barbican, since Authorization falls within Keystone's domain.


Tuesday November 4, 2014 17:30 - 18:10
Gauguin

17:30

Growth Challenges (Part 2)
After four years of continued expansion, the OpenStack project is encountering new growth challenges. Those call for rapid and aggressive solutions to ensure that we keep on growing and adapt to an ever-changing world in the future. In this cross-project workshop we'll explore various growth challenges (scaling core teams inside projects, scaling the gate, scaling the horizontal teams, how to get competition without duplication...), brainstorm solutions, and expose the current proposed reforms (big tent, layer 1...).



Tuesday November 4, 2014 17:30 - 18:10
Manet

17:30

Larger Policy Discussion
Etherpad: https://etherpad.openstack.org/p/kilo-crossproject-larger-policy-discussion

OpenStack's policy (rule engine and enforcement implementation) has served us reasonably well, but it is hard to work with and each deployment of OpenStack can have a very different rule-set on what a given role means. In order to provide for a better user experience in tools like the Dashbard, and to offer more fine-grained authorization abilities, we need to be able to answer the following things for any OpenStack deployment: What can I do with role X? To perform action Y, what role(s) do I need?

Session Leads: Morgan Fainberg, Nathan Kinder, Adam Young



Tuesday November 4, 2014 17:30 - 18:10
Degas

17:30

Manila Access Groups
One of the ideas that came out of the Atlanta summit was a proposal to add access groups to Manila. We've spent some time working on the design and would like to share it with the community. The implementation for it is far enough along that I could merge very early in Kilo.


Tuesday November 4, 2014 17:30 - 18:10
Corot

17:30

Kolla
Tuesday November 4, 2014 17:30 - 18:10
Dufy Le Meridien
 
Wednesday, November 5
 

09:00

Better Async Error Reporting in Cinder
Led by: Winston and Alex
A few 'special' create requests would by-pass scheduler in current Cinder work flow (i.e. API service directly sends requests to Volume service), such as creating snapshot, volume clone, volume from snaphot or volume replica. The intention for introducing such shortcut was because scheduler wasn't aware that such requests don't really need scheduling due to the fact that most storage backends doesn't support doing a snapshot of a volume that doesn't resides on it or create a clone of the volume on other backend. But in the cases where storage backend doesn't have enough capacity to serve the request, scheduler actually can be a good place to stop the request from going further.


Wednesday November 5, 2014 09:00 - 09:40
Derain

09:00

Exposing different capabilities in Ironic
Boot and deploy driver behavior differs in ways which are not exposed nor discoverable in the API. Such differences are a break in the abstraction layer that Ironic provides, representing either a deficiency in a driver or in our API.

Vendor drivers naturally differ in their capabilities. The touch point for such capabilities are hidden either within the vendor passthru API endpoint, or nestled within special node properties or driver_info which only that driver can interpret. Neither approach is discoverable by a client.

In a heterogeneous environment, how does Ironic expose these differences to Nova and to users?


Hint: it is possible to use node.properties.capabilities to expose hints to the Nova scheduler, but this is not standardized, nor automatic, nor well documented, nor easily interpretable by a direct user.


Wednesday November 5, 2014 09:00 - 09:40
Dufy Le Meridien

09:00

Keystone Object Lifecycle / Object Dependencies
The dependency injection that Keystone is currently using is not really a "dependency injection" pattern. This session is focused on improving the internal-to-Keystone object lifecycle (including dependencies).

* What should the dependency injection pattern in Keystone look like?
* What is the lifecycle of objects (controllers, managers, etc)?
* How do we deal with Optional dependencies?

Led by: David Stanek (dstanek)

Etherpad: kilo-keystone-object-lifecycle
Pre-Session Reading on Object Lifecycle and Dependencies


Wednesday November 5, 2014 09:00 - 09:40
Corot

09:00

Neutron Dev Process and Procedures
The opening session for Neutron will cover the following:

* Why is contributing to Neutron painful
* Subteam reorg
* Bug backlog
* specs vs. bugs and procedural changes
* Incubator vs. feature branch

https://etherpad.openstack.org/p/neutron-processes


Wednesday November 5, 2014 09:00 - 09:40
Degas

09:00

Cells
Despite being used in some large deployments, Nova cells functionality is largely ignored, unclear, and untested. Our first goal is to get it tested in the gate and we can touch on plans for that. Beyond that we would like discuss how we can integrate cells into Nova such that a default deployment is a single cell deployment. And we would like to discuss how we organize the code so that the normal flow includes cells, so developers don't ignore them.

Session moderator: Andrew Laski (andrew.laski@rackspace.com)

Required reading: ML post

Wednesday November 5, 2014 09:00 - 09:40
  • format json

09:00

Gap analysis for using Tempest in production
Tempest should be able to run against any OpenStack cloud as README says. However, it's difficult to run against production (not devstack) OpenStack cloud now. In this session, we will talk about the gap that we should resolve. For example, UX(there are hundreds configs..), stability, performance etc.. Topic for which we already have bp done/in progress/todo, e.g. Cleanup of resources, run with pre-provisioned accounts (no overall admin required for parallel run)

Some subtopics which may be discussed as part of this session are:

Tempest UX Improvements, revisiting a GUI/service, tempest/rally integration


Wednesday November 5, 2014 09:00 - 09:40
Gauguin

09:50

Cinder Automated discovery and storage config
Led by: Anjaneya "Reddy" Chagam
Currently admin enters storage information manually in cinder.conf. This is not a sustainable model in large enterprises that will have many different storage systems. Moreover it requires admin to have deep intelligence to identify storage system features and figure out best way to group them. Also solution providers would like to offer differentiated capabilities that admins can take advantage of during pool composition and volume provisioning.

* Propose adding discovery module with driver based framework with following changes:
** Database persistence for storage systems.
** REST APIs for storage systems, capabilities (create, update, delete, list operations).
** In order to accommodate legacy storage systems, ability to manually configure storage specs or additional behavior that admin deems useful.
** Changes to existing Cinder code to use database storage information instead of cinder.conf (backend portion only) e.g., volume startup, cinder-volume, scheduler changes.


Wednesday November 5, 2014 09:50 - 10:30
Derain

09:50

Infra: Infra-manual
Discuss how to make progress on the new manual for the project infrastructure. The manual is in three parts: developers, project leaders, administrators.



Wednesday November 5, 2014 09:50 - 10:30
Gauguin

09:50

Resource locking in Ironic
I wrote the TaskManager module to protect hardware operations from interruption in a distributed environment, but it's not perfect. A crashed Conductor can leave dangling locks that still require manual intervention to free; we haven't added automatic expiration to locks because some operations rightfully take a long time. Meanwhile, client requests get refused when a node is locked, regardless of what operation holds the lock. Even internal periodic jobs like sync_power_state take a lock, which can result in users getting unexpected NodeLocked errors.

Have we gone too far away from the original intent? Are we using locks too broadly, or do we need more fine-grained lock handling (types or reasons or duration-based locks)? Or should we simply make it easier for operators to manually release a lock, eg. by adding a REST verb to "unlock" a node?


Wednesday November 5, 2014 09:50 - 10:30
Dufy Le Meridien

09:50

Hierarchical Multitenancy
A followup session from the Juno cycle focusing on changes needed to authorization and policy for a hierarchical multitenancy enabled OpenStack. This session will also cover expansion of roles to be more fine-grained (handling the reseller use-cases).

Led by: Guang Yee (gyee), Raildo Mascena (raildo), Henry Nash (henrynash)

Etherpad: hierarchical-multitenancy-kilo-summit
Pre-session reading on the state of Hierarchical Multitenency and goals for Kilo.


Wednesday November 5, 2014 09:50 - 10:30
Corot

09:50

Neutron Split: Vendor Plugins and Drivers: Part 1
This covers the following:

* Neutron Plugins and Drivers: Where should they live?
* Common controller design patterns.
* Is DHCP a plugin for the purposes of this discussion?

https://etherpad.openstack.org/p/neutron-vendors


Wednesday November 5, 2014 09:50 - 10:30
Degas

09:50

Cells continued
... cells discussion continued


Wednesday November 5, 2014 09:50 - 10:30
  • format json

10:30

Break
Wednesday November 5, 2014 10:30 - 11:00
Developer lounge Le Meridien

11:00

Extracting Brick from Cinder
Led by: Walter Boring
The idea of 'brick' was created back in the Havana timeframe. All along it was meant to be a standalone library that Cinder, Nova and any other project in OpenStack could use. Currently brick lives in a directory inside of Cinder, which means that only Cinder can use it.

We want to extract the brick directory and encapsulate it into it's own pypi library that any python project can use.

So we need to do:

* First create a separate python project and release it into pypi
* * Add brick to cinder's requirements.txt
* * Remove the existing cinder/brick directory
* * Everywhere in Cinder that uses cinder/brick will need to be modified to use
* the new pypi library.


Wednesday November 5, 2014 11:00 - 11:40
Derain

11:00

Making Ironic Easier to Use
We all agree Ironic should be easier to use "stand alone". Let's nail down a timeline for some concrete improvements on this.

Relates to in-project functional testing.


Wednesday November 5, 2014 11:00 - 11:40
Dufy Le Meridien

11:00

Neutron Split: Vendor Plugins and Drivers: Part 2
This covers the following:

* Neutron Plugins and Drivers: Where should they live?
* Common controller design patterns.
* Is DHCP a plugin for the purposes of this discussion?

https://etherpad.openstack.org/p/neutron-vendors


Wednesday November 5, 2014 11:00 - 11:40
Degas

11:00

Nova Objects Status and Deadlines
We need to talk about where we are in the process of converting all the things to use Objects, as well as what our plans are for completing this work. The scheduler and resource tracker tasks are related to this, and we can discuss it at a high level here, but should reserve most of that for the scheduler sessions.

Session moderator: Dan Smith (dms@danplanet.com)


Wednesday November 5, 2014 11:00 - 11:40
Manet
  • format json

11:00

Oslo graduation schedule
Which modules are ready to graduate and what order do they need to be done?

https://etherpad.openstack.org/p/kilo-oslo-library-proposals



Wednesday November 5, 2014 11:00 - 11:40
Corot

11:00

Tempest scope in the brave new world
As part of the discussions during the Juno cycle about the gate scaling limits we've been encroaching on, one of the solutions was to switch a large portion of the functional testing burden back into the individual projects. So where does this leave tempest? This session will be concerned with defining exactly what the scope of tempest should be moving forward, and how it fits in the greater testing scheme within OpenStack.


Wednesday November 5, 2014 11:00 - 11:40
Gauguin

11:50

Cinder Rolling Upgrades
Led by: Duncan Thomas
The proposed change is multi-fold, and the implementation details are not 100% fleshed out yet.

1. All new RPC changes need to leave the old version in place and functional - this might mean inserting blank / default values into new fields, etc. How long the old version(s) need to live needs to be decided - I'd suggest at least a full stable release to enable upgrade from release to release.

2. When adding a new RPC, the code must behave sensibly if that RPC is not received on the far side - the state machine work might help with this since things can be timed out / retried. In some cases admin action might be required, e.g. state reset APIS. This will vary from change to change.

3. On startup, a manager must query the db and see if there are any RPC version requirements in the DB that apply to it. If the requirements cannot be met, then it should exit witha suitable log message. If they can be met, then the requirement should be cached to avoid having to query the DB again. An RPC can be added to cause the cache to be updated without restarting the manager, if desired, however that won't be in the initial version.

4. When sending an RPC, the manager should check the cache for the max version it is allowed to send, and use that. Again, this requires the code to inherently support this kind of fallback, which is a new requirement.

This should mean that an upgrade works as follows:

1. First signal (via a new RPC call or by restarting them) all managers to write their maximum supported RPC versions into the DB. This will be one record per RPC per manager.

2. Update one service. On startup, it works out the maximum RPC version it can currently send by looking at what other can receive and taking the minimum. It also updates the DB with what new versions it can handle, if any.

3. Rolling update more services. They update the DB as appropriate.

4. Either signal via RPC or restart all running services, they should now all see the new version RPCs are supported everywhere.


Wednesday November 5, 2014 11:50 - 12:30
Derain

11:50

Understanding the hardware we have
Even though we have not agreed [1] whether to call it introspection, interrogation, inventory, or analysis, there is general agreement that this functionality is essential to Ironic AND discrete from the process of locating new (unmanaged) hardware within a data center.

We had several competing proposals [2] for implementing this, and a proof-of-concept [3]. Given the intense interest in this feature, I would like this feature to be implemented during Kilo, but that requires a solution which accommodates both in- and out-of-band methods within a common API.


[1]
http://lists.openstack.org/pipermail/openstack-dev/2014-October/048810.html

[2]
https://review.openstack.org/125920
https://review.openstack.org/109088
https://review.openstack.org/108324
https://review.openstack.org/107344
https://review.openstack.org/103007
https://review.openstack.org/100951

[3]
https://github.com/Divius/ironic-discoverd


Wednesday November 5, 2014 11:50 - 12:30
Dufy Le Meridien

11:50

Neutron CLI and client lib
Covers the following:

* CLI and client library consistency synced with other projects.
* neutronclient options and command syntax
* Client library is very different from other projects
* Collaboration points with openstack-client and python sdk projects

https://etherpad.openstack.org/p/neutron-cli


Wednesday November 5, 2014 11:50 - 12:30
Degas

11:50

Nova: Nova/Glance client library
At the last summit we talked about refactoring many of our interactions with glance into a client library. This session is to track the status of that work.


Wednesday November 5, 2014 11:50 - 12:30
Manet
  • format json

11:50

oslo.messaging
- Where can we find more reviewers?
- What do we do to keep the drivers up to date? Should they move to their own source trees?
- How can we get python 3 support?
- What is the relationship between oslo.messaging and kombu?

https://etherpad.openstack.org/p/kilo-oslo-oslo.messaging


Wednesday November 5, 2014 11:50 - 12:30
Corot

11:50

Stable branches
In this session we'll talk about stable branch maintenance, and how to evolve our processes to support the continued growth of OpenStack. In particular we'll cover:

- Future stable/icehouse and stable/juno point release dates
- Icehouse will be supported for 15 months. How are we doing on that so far ?
- Decision on support timeframe for Juno
- State of stable gate and how to improve it
- Freeze enforcement and exception process codification
- How to handle growth and load
* Should we set up project-specific review teams ?
* Should we use liaisons ?
- Should we abandon using a specific ML, to increase team visbility ?
- Open discussion


Wednesday November 5, 2014 11:50 - 12:30
Gauguin

12:30

Lunch
Wednesday November 5, 2014 12:30 - 13:50
Le Bar Le Meridien

13:50

Objectify Cinder
Led by: Thang Pham
By using objects, we have a standardized interface to the database, as well as standardize the data passed between cinder services over RPC. This separates the code from the actual database implementation, making it easier for rolling upgrades.


Wednesday November 5, 2014 13:50 - 14:30
Derain

13:50

Image Conversion + OVF/OVA support -Split session
Image conversion:

What problem is this trying to solve?:

We've been discussing image introspection and image conversion for quite a bit. Unfortunately, we didn't have the base tools needed to implement this feature before but we do now. Now that async-workers has landed, we can start discussing how to take advantage of them and implement things like the 2 blueprints listed below.

Blueprints:
1. https://blueprints.launchpad.net/glance/+spec/introspection-of-images
2. https://blueprints.launchpad.net/glance/+spec/basic-import-conversion

etherpad:
https://etherpad.openstack.org/p/kilo-glance-image-conversion

---
OVF/OVA support:

What problem is this trying to solve?:

This is a follow up of the Julo mid-cycle meet-up discussion. As the on-going progress of artifacts, we'd discuss how to support the OVF/OVA based on the artifacts discussion.
There are several things that need be discussed, including the different OVF format versioning support and multiple-images-per-virtual-system, and how to co-operate to achieve this.

etherpad:
https://etherpad.openstack.org/p/kilo-glance-image-conversion


Notes:
As this is a split sessions, 20 mins are allocated to each topic.


Wednesday November 5, 2014 13:50 - 14:30
Dufy Le Meridien

13:50

Heat integration strategy, functional & unit tests
Discussion:

- Discuss plan for in-tree functional and integration tests, and relationship with tempest

- Making use of Rally for performance and scalability testing

- Our in-tree "unit" tests have, in some cases, totally insane and unmanageable amounts of coupling, such that one minor change to some code can result in completely unreasonable test fallout.

- Some desirable changes become impractcal or avoided purely due to test fallout

- Need to move towards less coupled actual unit tests, combined with some integration tests which are easier to maintain

- External projects Heat usage patterns (sahara, murano, tuskar)


Wednesday November 5, 2014 13:50 - 14:30
Degas

13:50

Infra: AFS
The infra team set up an AFS cell. Discuss whether and how we might use it.


Wednesday November 5, 2014 13:50 - 14:30
Gauguin

13:50

Nova: Unconference
This session is an opportunity for people to raise their specs which wouldn't otherwise be covered by the formal schedule for the summit.

Moderator: John Garbutt (john@johngarbutt.com)

Sign up here:
https://etherpad.openstack.org/p/kilo-nova-summit-unconference


Wednesday November 5, 2014 13:50 - 14:30
Manet
  • format json

13:50

A Common Quota Management Library
We have a few different groups that want to revamp quota management. What should a shared quota library look like, and how should we develop it?

https://etherpad.openstack.org/p/kilo-oslo-common-quota-library



Wednesday November 5, 2014 13:50 - 14:30
Corot

14:40

Cinder State Machine
Concurrent resource access in cinder is a problem that has caused resource corruption when simultaneous resources are mutated on by multiple cinder entrypoints (api and manager for example). In Icehouse there has been some addition & usage of locks around manager functions to queue up those requests when a resource is being simultaneous worked on by multiple functions (this stops one of those operations from concurrently mutating the underlying resource). Sadly this is more of a *sledgehammer* approach and hides the symptoms of the problem and makes it non-obvious when debugging what other requests are queued up behind the lock (or why dead-locking is occurring, if and when it does).

To help alleviate and hopefully solve this problem we will try to attack some of these issues in a different manner, integrating a *allowed* state transition table into the ``create_volume` workflow and doing *strategic* state transitions and aborting/erroring out when these state transitions are not allowed. In the future this will help create a concrete set of well defined states and transitions for other workflows as well (and will make it clear while looking at code and during debugging which transitions are allowed at the same time and what transitions are actively occurring).


Wednesday November 5, 2014 14:40 - 15:20
Derain

14:40

Catalog Indexing + Artifacts - Split session
Catalog Index Service:
The Catalog Indexing service is under discussion of whether or not it should be an optional API and be able to be enabled as a separate endpoint from the primary API service.

etherpad:
Catalog Index Service: https://etherpad.openstack.org/p/kilo-glance-catalog-index-service

---

Artifacs repository current implementation status, plans for Kilo and further improvements:

What problem is this trying to solve?:
1. Give a quick update on the current implementation status of Artifact repository:
1.1. Review proposed and implemented solution and architecture
1.2. Get feedback on Artifact Type plugin definition framework (so early adopters can start using it asap)
2. Check what is likely to land in K (additional changes to J spec, etc)
3. Discuss further (K+) plans and roadmap


etherpad:
Artifacts development: https://etherpad.openstack.org/p/kilo-glance-artifacts-current-state-of-development


Notes:
As this is a split session, 20 mins are allocated for each topic.


Wednesday November 5, 2014 14:40 - 15:20
Dufy Le Meridien

14:40

Heat Upgrades (testing, status, zero downtime)
Problem: What types of updates do we want to support? How can we achieve this?
Discussion:
Should we be looking at nova objects to handle rolling updates better? https://review.openstack.org/#/c/127532/2/specs/kilo/adopt-oslo-versionedobjects.rst
I have added a cross project topic also.

Versioning all-the-things (RPC, API)
(shardy) Not neccesarily saying I have answers to this or that I need to lead the session, but this has been playing on my mind lately..

So far we've taken a less than disciplined approach to versioning things:

- Adding stuff (and worse later having to remove or disable it) to ReST API endpoints without changing the API version, so the actual capabilities of a Heat endpoint are not easily discoverable for clients

- Changing RPC interfaces over time without ever bumping the RPC version, this causes major problems for those doing rolling live upgrades of multi-engine heat deployments for example.

- Also what about resource plugin interfaces, should these be versioned , as they're not tied to the template DSL version?


Wednesday November 5, 2014 14:40 - 15:20
Degas

14:40

Horizon Operators, Deployers and End Users
Come provide your feedback to the Horizon Team regarding:
* UX improvement suggestions
* Pain Points
* Missing Functionality
* What we have right


Wednesday November 5, 2014 14:40 - 15:20
Corot

14:40

Nova CI status checkpoint
What is currently working well with Nova CI? What could use improvement? What are our coverage gaps? This session will discuss the current state of Nova CI and the improvements we'd like to make in the Kilo cycle.

Moderator: Matt Riedemann (mriedem@linux.vnet.ibm.com)

Required reading:
https://etherpad.openstack.org/p/nova-ci-status-checkpoint-kilo


Wednesday November 5, 2014 14:40 - 15:20
Manet

14:40

QA/Infra: Gating relationships
Discuss evolving needs of project gates. Criteria on what project interfaces need what kind of testing or gating. How to ensure that is kept up to date over time. What can be automated? How to document the result so that developers understand what is being tested, how, and when.


Wednesday November 5, 2014 14:40 - 15:20
Gauguin

15:30

Cinder Scheduler To Support Over subscription
Led by: Xing Yang & Eoghan Glynn
'infinite' and 'unknown' were initially used for capacity reporting for vendors using thin provisioning. They are now given the lowest weighing and are not the recommended way of reporting capacities. The proposal is to include virtual capacity in capacity reporting and allow over subscription for thin provisioning. Virtual capacity "provisioned_capacity_gb" was already part of Winston's spec https://review.openstack.org/#/c/105190/6/specs/juno/volume-statistics-reporting.rst. This proposal adds support for "over subscription" in thin provisioning and will be built on top of Winston's spec.


* Backend reports virtual capacity in get_volume_stats, in addition to total capacity and free capacity.
* Scheduler checks available virtual capacity and available real capacity.
* Over subscription ratio will be used to control virtual capacity allocation.
* Used ratio will be used to control real capacity usage.
*These two ratios should prevent over provisioning.
* Send notification if capacity reaches upper limit controlled by over subscription ratio or used ratio.
* cinder spec: https://review.openstack.org/#/c/129342/


Wednesday November 5, 2014 15:30 - 16:10
Derain

15:30

Glance Modularity, different endpoints, etc.
Modularity, different endpoints, different processes, different service catalog entries:

What problem is this trying to solve?:
Pick back up the remaining discussion (from mini-summit) related to the above topic; partially from Artifacts and partially from Catalog Index service discussion.

etherpad:
https://etherpad.openstack.org/p/kilo-glance-modularity-endpoints-processes-catalog


Wednesday November 5, 2014 15:30 - 16:10
Dufy Le Meridien

15:30

Heat Convergence - resolving any questions
This is a placeholder to discuss the progress and issues found with the PoC work and help clarify some of the design. A lot more will get done in the Friday session, this is just a starting point to that.


Wednesday November 5, 2014 15:30 - 16:10
Degas

15:30

Horizon/Keystone Cross Topic Session
* Federation
* Policy
** Centralize Policy file into Keystone so that we don't need to store a copy in Horizon.
* Hierarchical Multi-tenancy
* Challenges around v3 as a default?
* Keystone API version discovery [from the ML topic]


Wednesday November 5, 2014 15:30 - 16:10
Corot

15:30

Virt Driver Consistency
We have previously discussed setting a minimum set of required
functionality for virt drivers. This helps to provide a base set of
expected functionality that Nova can provide, regardless of the
hypervisor being used. A set has been proposed, discussed, and it's
time for us to make some decisions in this space. This session will
endeavor to settle on those, or at least identify the steps required
to close on this.

Session moderator: Dan Smith (dms@danplanet.com)

Required reading:

Current feature matrix: https://wiki.openstack.org/wiki/HypervisorSupportMatrix
Proposed requirements: https://wiki.openstack.org/wiki/HypervisorSupportMatrix/Requirements


Wednesday November 5, 2014 15:30 - 16:10
Manet

15:30

QA and CI after merge
Moving forward there is a desire to reduce the scope of the gate to a stable and reliable "gold standard" and move the testing of special cloud configurations and scenarios to post merge. [0] This is to help reduce the complexity of the gate to make it more reliable and easier to debug for everyone. But, first we need to define what kind of scenarios are outside of the gold standard and should be tested post merge. There is also a gap in the tooling for this effort, we need a dashboard and other tools to advertise the test results properly and to simplify troubleshooting.

Discussion on test after code merge:
- Which features to verify after merge
- Which kind of tests (flacky, long running, multinode)
- How to track results trends
- How to save features and tests verified after merge from bit rot

[0] http://lists.openstack.org/pipermail/openstack-dev/2014-July/041057.html


Wednesday November 5, 2014 15:30 - 16:10
Gauguin

16:10

Break
Wednesday November 5, 2014 16:10 - 16:30
Developer lounge Le Meridien

16:30

Glance Supportability, logging and error codes
Glance supportability, logging and proposal of error codes:

What problem is this trying to solve?:

There was recently tweet from Justin Clayton (@justinclayton42) "If there was a video game called Sysadmin, the final boss fight would be OpenStack." This gives clear indication that OpenStack supportability is not on the level required for Enterprise level product.
What can we do to make sure that expanding Glance is not the reason OpenStack qualifying this title:

1) Further improvements how we log events in Glance and more clear log messages?
2) How can we improve our part of the documentation?
3) How could we attack our bugs more efficiently?
4) New/existing tooling for failure analysis? Should this be pan OpenStack (new project) or Glance specific?
5) Implementing unified error codes for exceptions raised and logging?
6) More reasonable OSprofiler trace points (https://github.com/stackforge/osprofiler)
7) Using Rally in Glance gates (it's alredy there) more efficiently? Probably it make sense to do some live demo of OSprofiler & Rally during summit.

etherpad:
https://etherpad.openstack.org/p/kilo-glance-logging-supportability-error-codes


Wednesday November 5, 2014 16:30 - 17:10
Dufy Le Meridien

16:30

Heat Autoscaling API
With the first stage of convergence, an Autoscaling API that supports multiple updates without queueing could well be feasible in Kilo. We need to discuss what work needs to happen to get the Autoscaling API back on track, and who is going to do it.

Discussion:
Possible steps
Rearrange class hierarchy
Implement separate DB tables (like e.g. software deployments)
Move implementations into a separate Python library (don't mix with resource plugins)
Add a ReST API


Wednesday November 5, 2014 16:30 - 17:10
Degas

16:30

Keystone-Horizon-CLI Federation/SSO
A federation and SSO focused session around the CLI and Horizon integration with the Keystone implementation (including Keystone-to-Keystone) federation. This session will be focused on both functionality and user-experience.

This is the 2nd part of the session, the first part is part of the Horizon Design Track

* What is the federated user / SSO workflow within Horizon?
* What is the federated user / SSO workflow within openstackclient?
* What changes are needed on the Keystone side to support these workflows?
* What changes are needed on the Horizon side to support these workflows?
* What changes are needed on the CLI (openstackclient) side to support these workflows?

Led by: Adam Young (ayoung), Steve Martinelli (stevemar), Marek Denis (marekd), David Lyle (david-lyle)

Etherpad: kilo-keystone-horizon-cli-federation-sso
Pre-session reading on the next steps for Federated Identity/SSO support in OpenStack.


Wednesday November 5, 2014 16:30 - 17:10
Corot

16:30

Nova Functional testing
While we have a way to test many behaviours in integration, when
things go horribly wrong, it's hard to figure out why, or reproduce.
We are therefore looking to add functional testing based on something
other than devstack to the Nova repository itself. This session will
discuss the problems we're seeking to solve, and how we can approach
that work in the Kilo cycle.

Moderator: Sean Dague (sean@dague.net)

Required reading:
https://etherpad.openstack.org/p/kilo-nova-functional-testing


Wednesday November 5, 2014 16:30 - 17:10
Manet

16:30

Sahara: Elastic Data Processing (EDP) futures
In this session we'll discuss further development of the Elastic Data Processing (EDP) feature in Kilo. We'll review the current state and direction and agree on initial plans.

Etherpad: https://etherpad.openstack.org/p/kilo-summit-sahara-edp


Wednesday November 5, 2014 16:30 - 17:10
Derain

16:30

TripleO Onramp and New Implementations
The on-ramp for TripleO can be a bit confusing and long. We'll discuss current efforts to clarify things and lower the barrier to entry for developers, testers, and users.

QuintupleO - OpenStack on OpenStack on OpenStack; This should be a brief summary of where the project is at and where we expect to be in Kilo.

Tuskar - A brief summary of the Tuskar status and short/long term goals.

Implementations - TripleO has been largely focused on a single implementation until now, but we have learned that many users are interested in using different tools for some or all of the parts of the system. We will ask each representative from each set of tools to summarize what tools they're using and where they expect to go with their involvement in TripleO for Kilo. Including, but not limited to, Puppet, Kolla (docker/kubernetes), and Ansible.


Wednesday November 5, 2014 16:30 - 17:10
Gauguin

17:20

Ceilometer Switchover to in-tree functional tests
Figure out the implications of switching over from Tempest coverage to the new project-specific functional testing approach.


Wednesday November 5, 2014 17:20 - 18:00
Dufy Le Meridien

17:20

Heat and Containers
Problem: Discuss possible approaches for container management with Heat
Discussion:
There are multiple options on how to manage containers (specifically docker) with heat.
At present there is an implementation proposed[1] to use software-config and instance docker hook along with a container-manifest - inline with google container manifest used in kubernetes[2]
However, this does not support lifecycle management of the pod/containers.
There is an alternate approach proposed by this [3] blueprint where k8s pod config resource, k8s deployment resource would be implemented. Inital spec proposed for this available here[4]. However, there are some issues as the k8s pod is not idempotent and there is still a need to write a hook.

[1] https://bugs.launchpad.net/heat-templates/+bug/1380886, https://review.openstack.org/#/c/128182/
[2] https://cloud.google.com/compute/docs/containers/container_vms.
[3] https://blueprints.launchpad.net/heat/+spec/k8s-container-pod-resources
[4] https://review.openstack.org/#/c/130117/


Wednesday November 5, 2014 17:20 - 18:00
Degas

17:20

Keystone Operators, Deployers, and DevOps
A focused session for the Keystone developers to get feedback on Keystone from Operators, Deployers, and the DevOps community.

Starting feedback questions:
* What feedback can you provide us on how we can best improve user experience?
* What feedback can you provide us on where the documentation is most lacking and how best to improve it?
* What integration scenarios are you facing with identity sources and federation?
* What are your biggest pain points around Keystone?

This session will start with but not be limited to the questions outlined above.

Led by: Henry Nash (henrynash), Nathan Kinder (nkinder), Brad Topol (topol)

Etherpad: kilo-keystone-operators-deployers-and-devops


Wednesday November 5, 2014 17:20 - 18:00
Corot

17:20

NFV proposed specs
OpenStack has an NFV sub-team that has been working on advancing
features in support of NFV use cases. The group collects use cases and
requirements from NFV operators and works to implement and track
blueprints in support of them.

The purpose of this session is to review specs for Nova that are
targeted at the Kilo release.

Moderator: Russell Bryant (rbryant@redhat.com )

Required reading:
http://etherpad.openstack.org/kilo-nova-nfv


Wednesday November 5, 2014 17:20 - 18:00
Manet

17:20

Tempest-lib moving forward
During the juno release the tempest-lib project was started. The goal of tempest-lib is to take some of the common glue in tempest and create an externally consumable library that will enable creating additional test suites easily using similar mechanisms to tempest. This session aims to be a discussion on how tempest-lib should evolve and what functionality should be in it as opposed to tempest.


Wednesday November 5, 2014 17:20 - 18:00
Gauguin

17:20

Sahara UX discussion
In this session, we will briefly review the current Data Processing UI in Horizon while pointing out things that can be improved upon. For the remainder of the time, we will propose work items and have discussions on design for workflow, screen layout, and the proper approach with respect to coding the enhancements.

Etherpad: https://etherpad.openstack.org/p/kilo-summit-sahara-ux


Wednesday November 5, 2014 17:20 - 18:00
Derain
 
Thursday, November 6
 

09:00

Bringing Gnocchi and Ceilometer together
Defining the plan during Kilo to improve Ceilometer and Gnocchi cooperation, including migration and co-existence issues.


Thursday November 6, 2014 09:00 - 09:40
Dufy Le Meridien

09:00

Heat Template format useability improvements
Problem: Some template authoring tasks are not obvious, difficult, or not possible
Discussion:
The point of this session would be to try and find/prioritise some useful small/medium features that improve the template writing experience.
https://etherpad.openstack.org/p/heat-useablity-improvements


Thursday November 6, 2014 09:00 - 09:40
Degas

09:00

Asserting hardware ready-state
Discuss a unified API for asserting ready state using either in-band and out-of-band means.

-----------

Hardware usually needs to be prepared before its first use. In many environments, the same process is repeated before each use. OnMetal already supports a "decom" phase (though somewhat of a misnomer, this is all about "ready-state").

That work has not landed upstream yet. However, the OnMetal team is giving a presentation on Wednesday about it:

http://openstacksummitnovember2014paris.sched.org/event/722245d15f368a720d95c9a9bbb77100


Thursday November 6, 2014 09:00 - 09:40
Gauguin

09:00

python-keystoneclient
This session is focused on the direction and improvements for python-keystoneclient within the Kilo cycle.

* Authorization Workflow: moving from username/password to token auth in a series of requests
* How should sessions/adapters be used when consuming a series of clients (e.g. python-cinderclient) from another python tool?
* Goals for supporting JSON-Home.
* What does the python-keystoneclient 1.0.0 release look like? What is the timeline to reach 1.0.0?

Led By: Jamie Lennox (jamielennox)

Etherpad: kilo-keystone-python-keystoneclient
Pre-session reading on the current state and direction of keystoneclient.


Thursday November 6, 2014 09:00 - 09:40
Corot

09:00

Nova API microversions
Now that we've agreed on a plan for a v2.1 API, we need to work out the specifics of how microversions will work.

Session moderator: Michael Still (mikal@stillhq.com)

Required reading:
https://review.openstack.org/#/c/127127 -- the microversion specification


Thursday November 6, 2014 09:00 - 09:40
Manet

09:00

Large and long-living clusters in Sahara
There are two attributes of Sahara clusters that impact manageability and automation: size and duration. Small, transient clusters are relatively easy to manage, but what about large and/or long-running clusters in a production environment? In this session we will explore issues related to large and long-running clusters and work on a development plan. Issues include but are not limited to high availability, partial deployment, retries during cluster launch, data locality management, and scratch directory cleanup.

Etherpad: https://etherpad.openstack.org/p/kilo-summit-sahara-production-clusters


Thursday November 6, 2014 09:00 - 09:40
Derain

09:00

Ops Summit: Hacking Documentation for Operators
Take some time from your Summit schedule to sit down with OpenStack documentation team members and learn how to improve OpenStack operator documentation. We'll have plenty of docs team members on hand so bring your laptops and doc bug lists and we'll hack together. Documentation doesn't have to be an afterthought or super difficult, we'll even take your crib notes in doc bug reports.

Moderators
avatar for Lana Brindley

Lana Brindley

Senior Manager, Information Development, Rackspace
Manager of a team of technical writers at Rackspace, and Liberty Documentation PTL.
avatar for Anne Gentle

Anne Gentle

Principal Engineer, Rackspace
Anne Gentle works in open source projects with the OpenStack project at Rackspace, using open source techniques for API design and documentation. She ensures the docs.openstack.org site contains relevant and accurate documentation for 20 projects written in Python across 130 git repositories. She advocates for cloud users and administrators by providing accurate technical information to increase OpenStack adoption as a cloud for the world... Read More →
avatar for Andreas Jaeger

Andreas Jaeger

SUSE
Andreas Jaeger is a core contributor to OpenStack documentation and the infrastructure CI (project-config). He works for SUSE as product manager. | | He has two kids and enjoy enjoys taking photos - not only of his family - and cycling. |
GP

Gauvain Pocentek

Objectif Libre

Thursday November 6, 2014 09:00 - 10:30
Hyatt - Palais Royal Hyatt Hotel

09:00

Ops Summit: Log Rationalisation Working Group
Working session for the Log Rationalisation Working Group, to take some action based on the feedback session on the Monday. All are welcome to attend to participate in the activity to make logging better in OpenStack.

https://etherpad.openstack.org/p/kilo-summit-ops-log-rationalisation

Moderators
avatar for Rochelle Grober

Rochelle Grober

Senior Staff Architect - Huawei, Huawei Technologies
Rocky is an industry veteran, with experience spanning computer bring up to AI, networks and embedded. But her attention always seems to return to close to the metal, large infrastructure. Starting out in EE, she migrated to SW development then on to QA and SW Process, which is why she is a champion for operations, interoperability and usability. She has been aware and peripherally involved in Open Source since the original creation of Copy Left... Read More →
avatar for Jay Pipes

Jay Pipes

Director of Upstream Engineering, Mirantis
Jay is Director of Upstream Engineering at Mirantis. He manages a team of senior engineers working on developing OpenStack projects in the community. Before Mirantis, he worked at AT&T, focusing on OpenStack development and operations. Previously, he was a Director of Engineering in HP s Open Source Cloud Services team, a development manager and software engineer at Rackspace Cloud, a software engineer at Sun Microsystems and the North American... Read More →

Thursday November 6, 2014 09:00 - 10:30
Hyatt - Vendome room Hyatt Hotel

09:00

Ops Summit: Telco Working Group
The first working session of the Telco Working Group! All telco folk and vendors welcome!

https://etherpad.openstack.org/p/kilo-summit-ops-telco

Moderators
avatar for Carol Barrett

Carol Barrett

Data Center Software Planner, Intel Corp
Carol is a 30+ yr high tech veteran, currently working within the Open Source Technology Center at Intel Corporation. She has broad software development experience ranging from weapon control systems on submarines to web based kids games to data center applications. She’s held many roles in her career including: Software Engineer, Engineering Manager, Business Development, Technology Marketing, Brand Management, Operations... Read More →
avatar for Stephen Gordon

Stephen Gordon

Principal Product Manager, Red Hat Canada
Stephen is a Sr. Technical Product Manager focused on OpenStack Compute and supporting technologies at Red Hat. Previously Stephen was a technical writer producing documentation for Red Hat Enterprise Linux OpenStack Platform, Red Hat Enterprise Virtualization, and related open source projects including the OpenStack documentation project. | | Stephen is an avid collector of “Internet points” at  the Foundation's Q&A portal... Read More →

Thursday November 6, 2014 09:00 - 10:30
Hyatt - Batignolles Hyatt Hotel

09:50

Mapping gnocchi semantics to InfluxDB/OpenTSDB
Revisit the gaps in InfluxDB in the light of new features added since mid-cycle and the emerging archive policy features in gnocchi, and consider the implications of getting Influx into the CI gate.

As for the OpenTSDB, we need some solution for the archiving policies and aggregation functions that will cover both TSDBs + Swift at least + OpenTSDB has no retention policies for now, although HBase has.


Thursday November 6, 2014 09:50 - 10:30
Dufy Le Meridien

09:50

Heat API and general useability improvements
Interacting with the Heat API via python-heatclient and diagnosing issues can be difficult.
This is part 2 of our useability sessions focused at improving our user experience.
https://etherpad.openstack.org/p/heat-useablity-improvements


Thursday November 6, 2014 09:50 - 10:30
Degas

09:50

Django-Angular Playing nice with Horizon
* How to handle serializing python objects
* How to handle fetching of directives
* How to handle internationalization using partial templates
* Clientside table rendering using concepts from above


Thursday November 6, 2014 09:50 - 10:30
Corot

09:50

nova-network migration
This session will cover migration plans from nova-network to neutron.

Moderator: Michael Still (mikal@stillhq.com)

Required reading:
http://lists.openstack.org/pipermail/openstack-dev/2014-August/044106.html
http://www.stillhq.com/openstack/juno/000014.html


Thursday November 6, 2014 09:50 - 10:30
Manet

09:50

Vulnerability Management
In this session we'll discuss the Vulnerability Management Team processes and how to make progress on them in Kilo. In particular, we'll cover:

- Publishing OSSAs to openstack/ossa in YAML, using Gerrit for signoff
- Create tooling to send emails to the right lists from merged OSSA YAML
- Automate publication to a security.openstack.org website
- Introduction of the VMT liaisons
- Incident Report taxonomy
- What projects would we support in a big-tent model
- Relationship with the OSSG
- Development of a severity metric
- Status of Gerrit for security patches


Thursday November 6, 2014 09:50 - 10:30
Gauguin

09:50

Running Sahara in production
For larger workloads and more critical use cases Sahara needs be robust and flexible in what it supports. In this session, we will discuss the existing gaps in Sahara that need to be improved upon to make Sahara a truly production-ready service. Topics covered are likely to include HA, distributed mode, upgrading Sahara, and bare metal clusters. What would *you* like to be able to do with Sahara?

Etherpad: https://etherpad.openstack.org/p/kilo-summit-sahara-production


Thursday November 6, 2014 09:50 - 10:30
Derain

10:30

Break
Thursday November 6, 2014 10:30 - 11:00
Developer lounge Le Meridien

11:00

Areas to target for collaboration with Monasca
Identify common functional areas such as alarming where we can share or converge implementations: alarming, anomaly detection, high-throughput messaging over Kafka for example.


Thursday November 6, 2014 11:00 - 11:40
Dufy Le Meridien

11:00

Horizon Clientside Data
Develop plan for Data loading and management clientside

Topics including:
* asynchronous data loading
* pagination/total count
* filtering
* sorting


Thursday November 6, 2014 11:00 - 11:40
Corot

11:00

Infra: Translation Tools
Meet with the translation team and work out a plan to migrate to a new translation system.



Thursday November 6, 2014 11:00 - 11:40
Gauguin

11:00

Neutron lightning talks
There is a huge focus on refactoring this cycle in Neutron. Our design summit is based around this. At the same time, we'd like to leave some time for people to propose other ideas and quickly share them with the community.

This slot is designed to have 6 lightning talks, each 5 minutes long.


Thursday November 6, 2014 11:00 - 11:40
Degas

11:00

Nova Scheduler and Resource Tracker
The community is agreed upon the need for a scheduler system that can take into account more than just compute resource usage calculations, however we have a significant amount of refactoring work that is needed before a split of the current Nova scheduler will be feasible. It is possible to get all that refactoring done, but we need a concerted effort.

This session is intended to finalize the plan for refactoring in Kilo, including:

* Sign-off on the foundational blueprints that turn currently unstructured interfaces into versioned, object-driven interfaces

* Isolating any calls to the Nova database from within the scheduler

* Stabilizing the RPC API in preparation for a split of the scheduler and an eventual REST API

As well as discussions around what pieces of data the scheduler should eventually own and changes proposed to the resource claim interfaces.

Etherpad:

https://etherpad.openstack.org/p/kilo-nova-scheduler-rt

Session moderator: Jay Pipes (jaypipes@gmail.com)

Required reading:

Active blueprints:

* Isolate scheduler DB:
https://review.openstack.org/#/c/89893/

* Detach service from compute node:
https://review.openstack.org/#/c/126895/

* Isolate scheduler DB for instance groups:
https://review.openstack.org/#/c/131553/

* Model resource objects:
https://review.openstack.org/#/c/127609/

* Model request spec object:
https://review.openstack.org/#/c/127610/

* Change select_destinations() to use RequestSpec object:
https://review.openstack.org/#/c/127612/


Thursday November 6, 2014 11:00 - 11:40
Manet

11:00

Ops Summit: Stable Branch
Moderators
avatar for Jonathan Proulx

Jonathan Proulx

Sr Technical Architect - Computer Science & Artificial Intelligence Lab, MIT

Thursday November 6, 2014 11:00 - 11:40
Hyatt - Vendome room Hyatt Hotel

11:00

Sahara's OpenStack integration and Security
(integration side)
We should continue looking at integration with other OpenStack services. This includes integration with Ironic for baremetal and hybrid data processing clusters provisioning. The list of notifications send to the Ceilometer should be discussed too.

(security side)
In this session we will discuss the current state of security in Sahara, emerging trends, and areas of improvement. The Juno release has seen the inclusion of two large security related features in the form of node level security groups, and Swift access for clusters through proxy users. One of our goals with regards to security for Kilo is the creation of a guidelines document including topics such as; best practices, suggested port lists for security groups, and in-depth discussions of current features.

Etherpad: https://etherpad.openstack.org/p/kilo-summit-sahara-integration-security


Thursday November 6, 2014 11:00 - 11:40
Derain

11:00

Ops Summit: Monitoring WG
A new working group to look at best practice for monitoring OpenStack

Moderators
avatar for Joe Topjian

Joe Topjian

Systems Architect, Cybera
Joe currently lives in Alberta, Canada, building and automating clouds for Cybera. Joe is also a co-author of the OpenStack Operations Guide and is an active member of the OpenStack Operators community.

Thursday November 6, 2014 11:00 - 12:30
Hyatt - Batignolles Hyatt Hotel

11:00

Ops Summit: Upgrades
A small working session on what we can do to improve the upgrade experience.

Moderators
avatar for Sam Morrison

Sam Morrison

Cloud Architect, NeCTAR, University of Melbourne
Sam Morrison is the Technical lead for the NeCTAR Research Cloud, which spans 9 different datacentres across Australia. In this role he is responsible for coordinating the technical team's activities in each location, in addition to OpenStack customisation work. As happy managing government funding milestone expectations as diving into python code, Sam is a contributor to nova, horizon, cinder, keystone, glance, ceilometer and trove. His latest... Read More →

Thursday November 6, 2014 11:00 - 12:30
Hyatt - Palais Royal Hyatt Hotel

11:50

Persistent pipeline config / Kafka publisher
This session covers two topics:

1. Store polling/notifications configurations into a durable data store for dynamic changes without service re-start

2. Discuss adding a new Ceilometer Publisher that will act as a Kafka Producer publishing ceilometer metrics to Kafka. This publisher can be provided as an option in the pipeline file along with the required Kafka configuration options such as the connection details, Kafka topic names.


Thursday November 6, 2014 11:50 - 12:30
Dufy Le Meridien

11:50

Neutron Advanced Services Spin Out
Covers the following in the context of existing *aaS things (LBaaS, VPNaaS, FWaaS):

* Lay a plan to spin services into their own git repository
* How does this affect the Tacker (service VM) project?
* How does this affect the flavors framework?

https://etherpad.openstack.org/p/neutron-services


Thursday November 6, 2014 11:50 - 12:30
Degas

11:50

Nova Scheduler and Resource Tracker continued
... continued discussion of scheduler and resource tracker work


Thursday November 6, 2014 11:50 - 12:30
Manet

11:50

Ops Summit: Chef
Chef users unite! Let's talk about cookbooks.

Moderators
avatar for JJ Asghar

JJ Asghar

Sr. Partner Engineer, CHEF
JJ is a Sr. Partner Engineer at Chef, he is also the PTL for the Openstack-Chef project. He lives in Austin, Texas and has been part of the OpenStack community since Diablo's release. | He enjoys a good strong stout, hoppy IPA, and some Dwarf Fortress. He's a member of the Church of Emacs, and attempts to push people to use Markdown over reStructuredText. | He's a father and husband, if he's not trying to automate his job away he's trying to... Read More →

Thursday November 6, 2014 11:50 - 12:30
Hyatt - Vendome room Hyatt Hotel

11:50

taskflow
How can we recruit more reviewers for taskflow? Should some of its features lib in other libraries?

https://etherpad.openstack.org/p/kilo-oslo-taskflow



Thursday November 6, 2014 11:50 - 12:30
Corot

11:50

Future Auth interface design for Tempest
As part of the work to stabilize the interfaces which should be externally consumable one of particular note is the auth layer in tempest. Currently it's designed to be fairly modular and has 2 implementations for either keystone v2 or v3, but it is still has some underlying assumptions about how the auth is setup. However, as part of the migration to tempest-lib this interface arguably should be created in a way which doesn't assume keystone is the auth mechanism being used. The in-tree implemenations of the interface will still be keystone specific, but it will allow external consumers to use whatever auth system they need.(for example swift auth) This session will discuss how we should implement the to-be auth interface in tempest-lib (by considering some use cases) to make it work for most use cases.


Thursday November 6, 2014 11:50 - 12:30
Gauguin

11:50

Swift: Erasure Code Overview
Currently many members of the community are building an erasure code storage policy into Swift. This session will review the current design and disuss the state of the feature.

http://specs.openstack.org/openstack/swift-specs/specs/swift/erasure_coding.html

We can be pretty dynamic about this, if something specific comes up at the time we can dive in but iniital thoughts are:
- overview of design spec
- deeper dive on MIME support, what it is and why we need it for PUT path
- reconstructor key concepts
- current status of development (or maybe we save this for open discussion)



Thursday November 6, 2014 11:50 - 12:30
Derain
  • format json

12:30

Lunch
Thursday November 6, 2014 12:30 - 13:40
Le Bar Le Meridien

13:40

Ceilometer Notifications standardization
Discuss making notification bodies standardized and versioned so that there can be contractual stability between producers and consumers.


Thursday November 6, 2014 13:40 - 14:20
Dufy Le Meridien

13:40

Neutron REST / RPC / Plugin API: Part 1
Covers the following:

* Refactoring the WSGI layer using pecan
* API microversioning
* API cross-project consistency
* nova parity from an API perspective
* Making API definitions dynamic enough to load completely external experimental extensions

https://etherpad.openstack.org/p/neutron-rest-api


Thursday November 6, 2014 13:40 - 14:20
Degas

13:40

Nova Unconference
This session is an opportunity for people to raise their specs which wouldn't otherwise be covered by the formal schedule for the summit.

Moderator: John Garbutt (john@johngarbutt.com)

Sign up here:
https://etherpad.openstack.org/p/kilo-nova-summit-unconference


Thursday November 6, 2014 13:40 - 14:20
Manet

13:40

Using alpha versioning for Oslo libraries
We use alpha version numbers on Oslo libraries to limit releases of new libraries to our development environment, in part to avoid risk in the stable maintenance builds. We also use them to indicate to distros that they may not need to package the alpha release but can use it for testing their packaging tools. There are a lot of objections to the use of alpha version numbers., and the main argument seems to be that we shouldn't be relying on alpha versions, which IMO ignores the fact that we need development and functional test time on new libraries just as we do with applications. This topic is then tied up with thes question of whether and how we do functional testing and unit cross-testing for libraries as well as the role of the incubator in evolving APIs.

https://etherpad.openstack.org/p/kilo-oslo-alpha-versioning



Thursday November 6, 2014 13:40 - 14:20
Corot

13:40

Swift Service tokens
This proposal originated in the desire to provide a more secure and
convenient way of storing service-specific data (such as Glance
images) in end-user Swift accounts. When disucssing this with the
Keystone team, the concept of composite authorization was invented.

https://review.openstack.org/#/c/105228/


Thursday November 6, 2014 13:40 - 14:20
Derain

13:40

TripleO - L3 Network Support / Cinder+Ceph HA
L3 - We want to be able to host baremetal machines on L3 segmented networks, as currently we are limited to a flat L2 configuration. There is work to be done in Nova and in our orchestration to support segmentation in installation. There will also likely need to be support in Tuskar for configuring these segments.

Cinder HA - We want to back Cinder with Ceph and have our Cinder services HA. Work has begun and should complete in Kilo, however there are some challenges left that we'll discuss.


Thursday November 6, 2014 13:40 - 14:20
Gauguin

13:40

Ops Summit: Ops Tools Working Group
Moderators
avatar for Andrew Mitry

Andrew Mitry

Distinguished Engineer, Walmart
Andrew Mitry is currently a cloud lead at Walmart. | | Andrew Mitry lead Comcast’s OpenStack team for the past three years where both the infrastructure and team have grown more than 500% in the past year. The team led Internet Protocol version 6 (IPv6) in OpenStack and ranked as a top 20 contributor with more than 36,000 lines of code.
avatar for Joe Topjian

Joe Topjian

Systems Architect, Cybera
Joe currently lives in Alberta, Canada, building and automating clouds for Cybera. Joe is also a co-author of the OpenStack Operations Guide and is an active member of the OpenStack Operators community.

Thursday November 6, 2014 13:40 - 15:00
Hyatt - Palais Royal Hyatt Hotel

13:40

Ops Summit: Blueprints Working Group
A continuation of the San Antonio discussions of the blueprints working group. If you are interested in helping operators work with blueprints, please attend and contribute to this session.

https://etherpad.openstack.org/p/kilo-summit-ops-blueprints

Moderators
avatar for Scott Devoid

Scott Devoid

Argonne National Lab
avatar for Jay Pipes

Jay Pipes

Director of Upstream Engineering, Mirantis
Jay is Director of Upstream Engineering at Mirantis. He manages a team of senior engineers working on developing OpenStack projects in the community. Before Mirantis, he worked at AT&T, focusing on OpenStack development and operations. Previously, he was a Director of Engineering in HP s Open Source Cloud Services team, a development manager and software engineer at Rackspace Cloud, a software engineer at Sun Microsystems and the North American... Read More →

Thursday November 6, 2014 13:40 - 15:50
Hyatt - Vendome room Hyatt Hotel

13:40

The Application Ecosystem Working Group
A working session of the Application Ecosystem Working Group, aiming to make real progress based on the feedback from Monday's session and look at the future work plan. All welcome!

https://etherpad.openstack.org/p/kilo-summit-ops-app-eco

Moderators
avatar for Chris C Kemp

Chris C Kemp

Founder of Nebula & OpenStack, Nebula
Chris founded and launched Nebula in July of 2011 to disrupt enterprise computing and ignite a new era of global innovation after serving as NASA's Chief Technology Officer. At NASA, Chris co-founded the OpenStack project and led the Nebula project, NASA's cloud computing initiative. Chris also served as the Chairman of the White House Federal Cloud Computing Standards Working Group and was a member of the Cloud Computing Executive Steering... Read More →
avatar for Everett Toews

Everett Toews

Developer Advocate, Rackspace
Everett Toews is a Docker Captain and a Developer Advocate at Rackspace making the cloud easy to use for developers and operators. Sometimes dev, sometimes ops, and sometimes advocate. He's interested in composing systems using APIs, SDKs, CLIs, and Docker. He likes to teach what he learns along the way.

Thursday November 6, 2014 13:40 - 15:50
Hyatt - Batignolles Hyatt Hotel

14:30

Authorization
Tokens have had a fair share of pain-points associated with them. This session is focused on addressing how we can move forward and improve the authorization mechanism that Keystone provides to OpenStack.

* How can UUID Tokens be improved (both functionality and user experience)?
* How can PKI/PKIZ Tokens be improved (both functionality and user experience)?
* Are there alternatives we should consider to the current token model?
* Other Improvements (Tokenless Operations, multiple-signers, etc).

Led by: Dolph Mathews (dolphm)

Etherpad: kilo-keystone-authorization
Pre-session reading on the current authorization mechanism.


Thursday November 6, 2014 14:30 - 15:10
Corot

14:30

Neutron REST / RPC / Plugin API: Part 2
Covers the following:

* More consistent RPC versioning toward live upgrade
* Improving the performance of RPC

https://etherpad.openstack.org/p/neutron-rpc-api


Thursday November 6, 2014 14:30 - 15:10
Degas

14:30

Nova Upgrades and DB migrations
We have made good progress towards no downtime upgrades. Lets make
sure we agree on how we move forward during kilo.

In particular, lets agree the order the services are upgraded, and
agree our approach for online DB migrations.

Session moderator: John Garbutt (john@johngarbutt.com)

Required reading:
https://review.openstack.org/#/c/102545/5/specs/juno/online-schema-changes.rst,cm
https://etherpad.openstack.org/p/kilo-nova-zero-downtime-upgrades


Thursday November 6, 2014 14:30 - 15:10
Manet

14:30

Devstack/grenade
Discussion of the direction that should be taken for devstack and grenade during the kilo cycle.


Thursday November 6, 2014 14:30 - 15:10
Gauguin

14:30

Swift Single-process performance optimizations
Discuss adding the ability to call storage servers directly from proxy when
all 4 servers are deployed in the same node. The conversation came up during the last
hackathon and we would like to continue during the Summit with a status of what's been done
and what the next steps are.


Thursday November 6, 2014 14:30 - 15:10
Derain

14:30

Building Out Trove Clusters
Now that we have a Clustering API, and an initial MongoDB implementation, let's discuss what are the next steps to merging other cluster implementations in Trove. Specifically we will look at the design for a galera cluster implementation (i.e. percona xtradbcluster) in Trove.



Thursday November 6, 2014 14:30 - 15:10
Dufy Le Meridien

15:10

Ops Summit: Puppet
Puppet users unite! Let's talk about openstack-puppet.

Moderators
avatar for Chris Hoge

Chris Hoge

Interop Engineer, OpenStack Foundation, OpenStack Foundation
Chris Hoge is the Interop Engineer for The OpenStack Foundation, focusing on interoperability testing and third party continuous integration testing. Formerly at Puppet Labs he lead the stackforge/puppet-openstack, one of the most popular tools for deploying and maintaining OpenStack deployments. He got his start in the OpenStack world operating a research Essex/Folsom installation at the University of Oregon for the College of Arts and... Read More →

Thursday November 6, 2014 15:10 - 15:50
Hyatt - Palais Royal Hyatt Hotel

15:20

Infra: Storyboard
Discuss progress and direction on storyboard and how to get involved.



Thursday November 6, 2014 15:20 - 16:00
Gauguin

15:20

Policy Model and User/Token Capabilities
Based upon the mailing list topic, this session is focused on what can be done with policy across all of OpenStack.

This is the more focused follow up session to Larger Policy Discussion Cross-Project Workshop

This session will include topics on:

* Centralizing policy.json information (in Keystone, in another project)
* What actions can a user perform with a given token? How is this information made available (API)?
* What roles are needed to perform a specific action? How is this information made available (API)?
* Constraints enforced for a given token (proposed spec: https://review.openstack.org/#/c/123726/ )

Led by: Morgan Fainberg (morganfainberg), Adam Young (ayoung), Nathan Kinder (nkinder)

Etherpad: kilo-keystone-policy-model-token-capabilities
Pre-session reading on the current policy model and goals of this session.


Thursday November 6, 2014 15:20 - 16:00
Corot

15:20

Neutron REST / RPC / Plugin API: Part 3
Covers the following:

* Re-architecting the base plugin
* Composition over mixins
* Supporting vendor-specifc features

https://etherpad.openstack.org/p/neutron-plugin-api


Thursday November 6, 2014 15:20 - 16:00
Degas

15:20

Containers service (magnum)
Magnum is conceived as two main components: An API service and an agent that will run in a Nova Instance to allow for container management. We think that the API should closely track the API for Docker, plus async capability and multi-tenancy with python-keystone integration. The agent will use Zaqar queues to accept commands from the Magnum API service. Each tenant will own a unique Zaqar queue, and agents will listen on the queue for instructions. This session will provide an overview of the architectural plan, and we will discuss the design of the agent.

Session moderator: Adrian Otto (adrian.otto@rackspace.com)

Required reading:
https://review.openstack.org/114044 -- the Magnum specification
https://blueprints.launchpad.net/magnum/+spec/magnum-agent-for-nova


Thursday November 6, 2014 15:20 - 16:00
Manet

15:20

Swift On-Disk Encryption
Update/Status/Next steps on work being done related to at-rest encryption in Swift.

See https://review.openstack.org/#/c/123220/


Thursday November 6, 2014 15:20 - 16:00
Derain

15:20

Replication v2 and Scheduling Tasks in Trove
Juno had a v1 implementation of Replication for MySQL. During the Atlanta design sessions we also began a discussion of the framework for scheduling tasks. During this design session we will discuss what the next steps are for replication, and scheduled tasks in Trove.


Thursday November 6, 2014 15:20 - 16:00
Dufy Le Meridien

16:00

Break
Thursday November 6, 2014 16:00 - 16:30
Developer lounge Le Meridien

16:30

Making Horizon more easy to customize and extend
Possible Topics:
* customize table row actions
* customize table tabs
* customize permissions/policy_rules (without touching source code) for most of interface objects (currently only Panels/Dashboards/Actions are affected). Proposed additional types of objects:
** workflow steps
** merging permissions, policy_rules and can_access?
* Customise Angular


Thursday November 6, 2014 16:30 - 17:10
Corot

16:30

Neutron Pluggable IPAM
Covers the following:

* Pluggable IPAM system in Neutron
* Scheduler items related to this
* Possibly may touch on DHCP agent issues

https://etherpad.openstack.org/p/neutron-ipam


Thursday November 6, 2014 16:30 - 17:10
Degas

16:30

Setting nova's roadmap for Kilo
As the last session of the summit, we will work on a nova roadmap that will cover the most important things for nova's future, and pick the top few as project priorities for Kilo.

Session moderator: Joe Gordon (joe.gordon0@gmail.com)

Required reading:
http://docs.openstack.org/developer/nova/devref/kilo.blueprints.html#project-priorities
http://specs.openstack.org/openstack/nova-specs/priorities/kilo-priorities.html
https://etherpad.openstack.org/p/kilo-nova-priorities


Thursday November 6, 2014 16:30 - 17:10
Manet

16:30

Python 3 support in Oslo
Status, blockers, and plans for work in kilo.

https://etherpad.openstack.org/p/kilo-oslo-python-3



Thursday November 6, 2014 16:30 - 17:10
Gauguin

16:30

Status of python-swiftclient
Most of us agree that python-swiftclient is pretty bad to both use and maintain. However, this is the first experience that many clients have with Swift, and it gives a bad impression. In this sessions, let's cover the current state and make some plans on how to make swiftclient awesome


Thursday November 6, 2014 16:30 - 17:10
Derain

16:30

Testing Trove
How can we improve testability of the various aspects, and components
of Trove. Specifically, this would cover topics including the likes of:

* How do we test clustering?
* How do we test third-party datastores?
* The death of the deprecated Trove CI
* Grenade testing for trove


Thursday November 6, 2014 16:30 - 17:10
Dufy Le Meridien

16:30

API Working Group

Goal: Record pain points in the APIs and discuss solutions Etherpad: https://etherpad.openstack.org/p/kilo-crossproject-api-wg
Standardizing response codes
  • We should have proposals already submitted to the openstack/api-wg repo
  • Discuss/debate anything that isn't consensus already
API Extensions
  • How to gracefully move away from them?

Operator vs. User APIs
  • Do we recommend splitting them out?

Discoverability
  • For the route map
    • JSONHome is a good choice here
  • For the request and response payload
    • JSONSchema+HAL is a good choice here
Documentation for the API itself
  • Swagger 2.0

Plans for pagination/filtering/sort/count

Moderators
avatar for Jay Pipes

Jay Pipes

Director of Upstream Engineering, Mirantis
Jay is Director of Upstream Engineering at Mirantis. He manages a team of senior engineers working on developing OpenStack projects in the community. Before Mirantis, he worked at AT&T, focusing on OpenStack development and operations. Previously, he was a Director of Engineering in HP s Open Source Cloud Services team, a development manager and software engineer at Rackspace Cloud, a software engineer at Sun Microsystems and the North American... Read More →
avatar for Everett Toews

Everett Toews

Developer Advocate, Rackspace
Everett Toews is a Docker Captain and a Developer Advocate at Rackspace making the cloud easy to use for developers and operators. Sometimes dev, sometimes ops, and sometimes advocate. He's interested in composing systems using APIs, SDKs, CLIs, and Docker. He likes to teach what he learns along the way.

Thursday November 6, 2014 16:30 - 18:00
Hyatt - Vendome room Hyatt Hotel

16:30

Ops Summit: Large Deployments Team
First working session for the Large scale Deployments Team!

https://etherpad.openstack.org/p/kilo-summit-ops-large-deployments

Moderators
avatar for Matt Van Winkle

Matt Van Winkle

Senior Manager, Operations, Rackspace
I serve an amazingly talented group responsible for running the Rackspace public cloud. I have been with the company in various roles for over 13 years.  My primary task is helping my teams of engineers and developers build the software that manages our constantly growing fleet of servers, enables easier day to day operations for our 24X7 Ops teams and facilitates smoother deployments by our software teams working on upstream OpenStack and... Read More →

Thursday November 6, 2014 16:30 - 18:00
Hyatt - Palais Royal Hyatt Hotel

16:30

Ops Summit: Win The Enterprise Working Group
Working Session for the Win The Enterprise Working Group 

https://etherpad.openstack.org/p/WTE_2.0

 



Moderators
avatar for Carol Barrett

Carol Barrett

Data Center Software Planner, Intel Corp
Carol is a 30+ yr high tech veteran, currently working within the Open Source Technology Center at Intel Corporation. She has broad software development experience ranging from weapon control systems on submarines to web based kids games to data center applications. She’s held many roles in her career including: Software Engineer, Engineering Manager, Business Development, Technology Marketing, Brand Management, Operations... Read More →

Thursday November 6, 2014 16:30 - 18:00
Hyatt - Batignolles Hyatt Hotel

17:20

UX and Horizon
How does UX fit into the Horizon development process?

What are some specific areas of improvement?

Specific work items:
* Launch Instance
* Network Creation
* Detail Views


Thursday November 6, 2014 17:20 - 18:00
Corot

17:20

Paying down technical debt in neutron agents
Covers the following:

* L2 Agent refactor
* L3 Agent refactor
* DVR expansion to more than OVS

https://etherpad.openstack.org/p/neutron-l2
https://etherpad.openstack.org/p/neutron-l3


Thursday November 6, 2014 17:20 - 18:00
Degas

17:20

Setting nova's roadmap for Kilo continued
... continued discussion of the kilo roadmap


Thursday November 6, 2014 17:20 - 18:00
Manet

17:20

Moving Oslo away from namespace packages
Should we? How do we do it?

https://etherpad.openstack.org/p/kilo-oslo-namespace-packages



Thursday November 6, 2014 17:20 - 18:00
Gauguin

17:20

Swift Ops feedback session
Feedback session from people running Swift clusters in production


Thursday November 6, 2014 17:20 - 18:00
Derain

17:20

Trove integration with other OpenStack components
We're doing a fair bit of work integrating with oslo-modules, and
removing defunct incubator code. In the same vein, let's discuss areas in which Trove can better integrate with other OpenStack projects out there:

* OpenStack client for integration with the common client
* Other oslo modules
* Ceilometer
* Heat
* Barbican for Security
* pecan/wsme


Thursday November 6, 2014 17:20 - 18:00
Dufy Le Meridien
 
Friday, November 7
 

09:00

Ceilometer contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 09:00 - 12:30
Derain

09:00

Cinder contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 09:00 - 12:30
Manet

09:00

Heat contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 09:00 - 12:30
Corot

09:00

Infra / QA / RelMgt contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 09:00 - 12:30
Gauguin

09:00

Neutron contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 09:00 - 12:30
Neuilly lobby Le Meridien

09:00

Nova: Work session
Work sessions are for Nova contributors to discuss implementation details and making quick progress over specific issues, in a small work group environment.

Friday November 7, 2014 09:00 - 12:30
Dufy Le Meridien
  • format json

09:00

Sahara contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 09:00 - 12:30
Derain

09:00

Swift contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 09:00 - 12:30
Manet

09:00

TripleO contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 09:00 - 12:30
Degas

09:00

Trove contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 09:00 - 12:30
Degas

12:30

Lunch
Friday November 7, 2014 12:30 - 13:40
Le Bar Le Meridien

13:40

Ceilometer contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 13:40 - 17:10
Derain

13:40

Glance contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 13:40 - 17:10
Derain

13:40

Heat contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 13:40 - 17:10
Corot

13:40

Horizon contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 13:40 - 17:10
Manet

13:40

Infra / QA / RelMgt contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 13:40 - 17:10
Gauguin

13:40

Ironic contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 13:40 - 17:10
Degas

13:40

Keystone contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 13:40 - 17:10
Manet

13:40

Neutron contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 13:40 - 17:10
Neuilly lobby Le Meridien

13:40

Nova contributors meetup
The Nova contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 13:40 - 17:10
Dufy Le Meridien
  • format json

13:40

TripleO contributors meetup
The contributors meetup is a informal gathering of the project contributors, with an open agenda.

Friday November 7, 2014 13:40 - 17:10
Degas

17:20

Design Summit feedback
In this closing session, we'll reflect back on how this Design Summit went and how we can improve in the future.

"Design Summits" are a key moment in OpenStack development cycles where our development community gathers to brainstorm the next steps for our project, it's critical that we get it right and continuously improve on the format.

We introduced contributors meetups on the Friday this time around: was that a success ? Should we do them again ?

Friday November 7, 2014 17:20 - 18:00
Manet