2017 Events

Speaker: Steven Ellis

We are potentially at an interesting tipping point in the future of Free and Open Source software as more of our interactions are managed by services rather than applications. Does Open Source have a future or are we pivoting towards a services based ecosystem where the best we can hope for is some degree of Open Standards? These open standards have no specific entity tasked with defining each of them, yet every day we hear about Open Hardware, Open Government, Open APIs.What does Open mean within the ecosystem you operate within, and how can we continue to be a catalyst for the journey and grow the Open Source movement when so many closed outcomes are being produced from the Open Source ecosystem?This is a collaborative session to look unpack the issue and formulate some actions that can be used by the attendees within their communities, businesses and organisations.

Speaker: Steven Pousty

Everybody cares about the place (they live, they grew up in, they had a great vacation, in the news….). With the rise of open data, big data tooling, and new visualisation technology, we can actually now build applications that give people new ways to explore beyond “where is the closest Starbucks”. I have collected Open Data from my home town (Santa Cruz, CA) and compiled it into the beginnings of a visualization and analysis platform. The goal of this talk is to show the process of collecting open data from disparate sources, some of the caveats on being able to put them together, general lessons learned, and some fun visualizations. I want to move past thinking about sources for open data and moving on to tools and lessons so you can get cracking! I want to show how we can enable people to gather open data and turn it to open knowledge. Data sources will be from Government (e.g. United States Geologic Survey) and Non-Government sources (e.g Audubon Society eBird Data) while some of the tools covered will be Apache Spark, PostGIS, Leaflet, and various others.

Speaker: Steven Ellis

Tips and tricks for both operators and developers to get the most out of ThinLVM and KVM when building demo environments, or optimising real world business solutions.

Speaker: Rikki Endsley

Are you sick of seeing job listings calling for “rock star developers”? What does that even mean? Developers who get the glory, while the band, agent, road crew, and sound engineers do the work? Instead of being a one-hit wonder who crashes and burns by 27, look to Willie Nelson for inspiration. Willie’s 60+ years in the music business offer plenty of lessons developers can apply to their own careers. Attend this fun talk to see how collaborating with a diverse mix of peo ple, learning new skills, choosing the best tool for the job (and then improving on it), contributing to a range of communities, and not being afraid to fail have benefited Willie’s career. The talk will be fun, but the lessons are practical for developers–and people who work with them–of all experience levels.

Speaker: Jason Shepherd

Recent research by Chris Frohoff and Gabriel Lawrence has exposed gadget chains in various libraries that allow code to be executed during object deserialization in Java. They’ve done some excellent research, including publishing some code that allows anyone to serialize a malicious payload that when deserialized runs the operating system command of their choice, as the user which started the Java Virtual Machine (JVM). The vulnerabilities are not with the gadget chains themselves but with the code that deserializes them.There are couple of ways in which this type of attack on the JVM can be mitigated:

• not deserializing untrusted objects;
• not having the classes used in the ‘gadget chain’ in the classpath;
• running the JVM as a non-root operating system user, with reduced privileges;
• egress filtering not allowing any outbound traffic other than that matching a connection for which the firewall already has an existing state table entry.In this talk we’ll explore each of these in detail, using JBoss EAP as a example of how these mitigations can be introduced.

Speaker: Fraser Tweedale

This turbo talk will cover advances in FreeIPA over the past two years

Speaker: Rikki Endsley

Whether you’re a layperson talking about free software law, or a lawyer speaking to a layperson, how you deliver your message determines how well your audience understands it. In this talk we’ll look at common mistakes laypeople make when interpreting legal issues, and tips for legal experts for condensing and explaining complicated topics to the layperson. We’ll also take a quick look at hot free software law and policy topics, the kinds of people writing about them (e.g., lawyers vs. programmers), and what’s missing (i.e., the topics community members want to learn more about).

Speaker: Richard Fontana

During the past few years, some have argued that we are living in a “post-open source software” (“POSS”) era, characterized by a new generation of developers who work on public software projects without any effort to indicate licensing or any regard to project governance. One view is that this represents an increase in carelessness or legal misunderstanding among developers. However, others have suggested that this behavior may be a deliberate reaction against the “permission culture” inherent in traditional approaches to open source. GitHub has been blamed for this phenomenon and it has taken some measures to address it. This talk will explore some legal and policy issues surrounding the tendency (real or supposed) for developers to fail to explicitly license their code

Speaker: Nick Coghlan

For a long time, the preferred approach to network service security has been the hardened bunker: define a system, deploy it, and patch it promptly when security vulnerabilities in the components used are reported and fixed.However, continuous integration services and automated deployments have opened up a new model: the moving target that grabs new releases of dependencies almost as soon as they are available, runs them through the CI process like any other software change, and then deploys them to production. Even if a new security flaw slips through testing, that’s considered better than being exposed to the flaws that were classified as normal bug fixes, but actually represented security issues.release-monitoring.org is a shared community service born out of the Fedora Infrastructure team that monitors for new upstream releases, and emits structured events that can be used to automatically trigger appropriate follow-on action.So if you’d like to learn how to drink from the firehose, this talk’s for you!

Speaker: David Airlie

OpenGL has long been the 3D API used on the Linux platform. However recently the Khronos group released the Vulkan 1.0 standard which covers a new closer to the metal API for 3D graphics programming across multiple operating systems. This talk will provide a brief introduction to the API and how supported the API is under Linux. The speaker will also discuss possible future uses for this new API and the advantages it brings.

Speaker: William Brown

Rust is a modern language developed by Mozilla pursuing the trifecta: Safety, Speed, and Concurrency. With such promise for the future, how can we use this now? We are not always in the beautiful open green fields; we must contend with our existing grunty applications!389 Directory Server is based on code now more than 20 years old. We cannot throw this out and replace it, but we want the benefits Rust gives us - especially for authentication and security critical code.We will explore the challenges of security in engineering, and behaviours of the modern programmer. I will discuss why we have spiraled down a mountain of failure as an engineering discipline - and why we need tools like Rust to validate our work. I will show faults in Directory Server that could have been prevented with safe language techniques. Finally, I will explore the Directory Server plugin interface, where we have the ability to provide pure Rust plugins, allowing safe, fast extensibility to a core piece of systems authentication. This will explore the challenges to build it, and the design patterns needed to make sure that the rewrite of the application in Rust is possible, today.

Speaker: David Chinner

After I ran out of talent and had a high speed encounter with an immovable object in May 2016, I needed to build a new race car. The car I crashed is somewhat unique for many reasons - it’s a Locost Clubman. The car has a space frame chassis, based on the original 1957 Lotus 7 designed by Colin Chapman. In the early 1990s, Ron Champion published a Haynes manual titled “Build your own sports car for as little as £250” which was a step by step guide to fabricating and building the entire car. Essentially, it was the source code for the car.Are the open source 3D cad tools capable of building a complex space frame chassis? Can I get an open source finite element analysis software to perform strength and failure analysis and get sane results? Can I design all the parts I need and send them to 3rd parties to get them laser cut, bent, fabricated and delivered to my door? Can I integrate my little CNC milling machine into these tools (i.e. CAM) so I can rapidly fabricate prototypes and iterate designs? If the open source tools exist, is it even possible to make them work together in a sane way? What tools have I had to write myself? What landmines will I step on?

Speaker: Rikki Endsley

In 2016, I adopted my first carnivorous plant, a Venus Fly Trap, which I named Gordon. I quickly discovered that, in addition to his ability to catch the occasional bug and get energy from the sun, Gordon’s health was closely connected to the environment and care I provided. In this talk, I’ll draw from more than a decade of experience working with open source communities (and a few months of experience keeping Gordon alive) to explain how caring for a Venus Fly Trap is much like caring for a community. Like carnivorous plants, communities can’t be left alone to fend for themselves, and they won’t explicitly tell you when there is a problem. Communities and carnivorous plants need an energy source, healthy environments, pest control, and occasional maintenance. Whether you’re a seasoned community gardener, or just starting to develop your carnivorous plant green thumb, you’ll learn practical tips for nurturing healthy roots that grow thriving communities (… or Venus Fly Traps).

Speaker: Tom Callaway

Tom Callaway, the Fedora Legal chair, will talk about the past,present, and future of licensing and legal issues in the Fedoracommunity. Tom is not a lawyer, nor does he play one on TV, but he doesconsult with lawyers, and occasionally, go drinking with them. Bringyour questions, and he’ll do his best to answer them. I am not alawyer, so nothing in my presentation should be (or could be) construedas legal advice.

Speaker: Mario Torre

Welcome to the Free Java DevRoom!

Speaker: Severin Gehwolf

Thermostat is a monitoring and management tool for Java deployments,allowing users to measure and monitor a host of different performanceaspects of their Java applications. Available metrics range from rawCPU and memory usage to operation of the Garbage Collector and JITcompiler through to thread activity and method call/heap profiles.Thermostat provides a GUI view of activity of local and distributedJVMs in a live-view or, alternatively, offline for after-the-factanalysis.

What Thermostat cannot do on its own is track events and recordstatistics that are specific to a given Java application, at least notunless the application co-operates with it, for example by publishingJMX statistics that Thermostat can read, persist and display in itsGUI. However, that’s about to change thanks to work Thermostatdevelopers have been doing to integrate Byteman into Thermostat.

Byteman is a tool which can be used to modify the behaviour of Javaprograms by injecting extra Java code almost anywhere in the program.You don’t need to recompile your program or even prepare it in advancein order for this to work. You can specify changes to the program onthe command line but, what is more amazing, you can actually useByteman to change the way a program runs after startup while it isstill running.

In this talk we will show how Thermostat can collect and visualizemetrics with Byteman’s help in order to better understand a specificperformance issue in a Java application.

Speaker: Christine H Flood, Roman Kennke

Garbage Collection (GC) liberates the programmer from having to callmalloc and free. More importantly GC saves the programmer from havingto debug their mistakes when using malloc and free. Unfortunately thedetails of how GC works are often a black box. This talk will startwith a tour of all of the GC algorithms currently available in OpenJDK.We’ll discuss how they work, their strengths and weaknesses, and whichclass of applications they were developed for. We’ll work our waythrough serial gc, parallel gc, concurrent mark and sweep, and g1.We’ll make the case for why we need all of them and just one more GCalgorithm, Shenandoah.

Shenandoah is a parallel and concurrent GC algorithm designed forapplications with 100gb+ heaps and tight pause time constraints. It’sthe first GC algorithm targeting OpenJDK which compacts the liveobjects while the Java threads are running. We’ll describe thealgorithm itself, the implementation details, and the optimizationsneeded to achieve good performance. We’ll present performance numbersand give a demo that visualizes Shenandoah.

Speaker: Charles Nutter

Ruby’s Strings aggregate a collection of bytes and an encoding,allowing for IO to avoid transcoding, regular expressions to executeagainst raw bytes, and 7-bit strings to be compactly represented. Onlythe last item has been adopted by Java. To make matters worse, mostJava APIs depend on Java’s string representation, making themincompatible with alternative languages like Ruby. We’ll explore theadvantages of Ruby’s string compared to Java’s and discuss options forimproving Java’s string support in the future.

Speaker: Christine H Flood

Speaker: Mark Reinhold, Mario Torre, Andrew Haley, Georges Saab, Doug Lea

An open Q&A session with members of the OpenJDK Governing Board

Speaker: Mark Wielaard

gcc/glibc support fortification of some functions by defining_FORTIFY_SOURCE. This inserts some compile and runtime buffer overflowchecks for selected glibc functions. These checks have no or verylittle runtime overhead and work on the object level (the compilerprovides/proofs the size of the object buffer size). valgrind memcheckprovides similar memory buffer overflow checks. These checks don’t needany compiler help (you won’t have to rebuild your code). But they havea much higher runtime overhead. They also work on a different level.valgrind memcheck doesn’t know anything about the objects the user ismanipulation but has knowledge of all memory blocks allocated. Letsexplore how these different mechanisms work and how we can make themwork better together.

Speaker: Mark Wielaard

Come and hack on Valgrind together. Open discussion about small (orbig) ideas to improve or change Valgrind.

Valgrind developers and users are encouraged to participate either bysubmitting ideas/suggestions or by joining the discussion. And ofcourse by kindly (or bitterly) complain about bugs you find importantthat are still Not YET solved for that many years!?@!!!

Afterwards we will sit together and try to fix or implement some of thethings discussed.

Speaker: Yaacov Zamir

There are many IoT dashboards out on the web, most will require networkconnection to a server far far away, and use non standard protocols. Wewill show how to combine free software tools and protocols from theworlds of IT monitoring, Industrial control and IoT to create simpleyet robust dashboards.

Modbus [1] is a serial communication protocol developed in 1979 for usewith programmable logic controllers (PLCs). In simple terms, it is amethod used for transmitting information over serial lines betweenelectronic devices., it’s openly published, royalty-free, simple androbust.

Many industrial controllers can speak Modbus, we can also teach “hobby”devices like Arduino boards and ESP8266 to speak Modbus [2]. Reliable,robust and simple free software Modbus client [3] will be used toacquire the metrics from our device, then the metrics will be collected[6][7] and sent to Hawkular and Grafana [8] to store and visualize ourdata.

(*) http://starwars.wikia.com/wiki/Bocce

[1] https://en.wikipedia.org/wiki/Modbus [2]https://github.com/yaacov/ArduinoModbusSlave [3]https://github.com/yaacov/node-modbus-serial [6]https://github.com/hawkular/hawkular-client-python [7]https://github.com/yaacov/hawkular-client-cli [8]https://github.com/hawkular/hawkular-grafana-datasource

Speaker: Lorenzo Mangani

HOMER 6 allows users to Export, Analyze and Alert RTC and VoIP sessionsin real time using popular Big-Data backends such as InfluxDB andElasticsearch providing unprecedented flexibility and opening the wayfor new uses of the platform in larger ecosystems with businessintelligence feeds.

Speaker: Charles Nutter, Thomas Enebo

JRuby has been compatible with various Ruby versions during itslifespan, ranging from the 1.6 series through today’s 2.4. JRuby hassupported Rails in some capacity since the 1.0 days. And at the sametime, we’ve continued to improve performance. In this talk we’llexplore JRuby’s level of compatibility today and discuss the challengesof keeping up with an actively-developed language and ecosystem.

Speaker: Stephan Bergmann

“But what about the extensions?” can be the death knell for whatevercool new feature somebody tries to implement in LibreOffice, asextensions naturally ask for interface stability. But what about them,anyway? Are they the saviour that brings diversity to our desktops, orare they just a ghostly phantom that stifles innovation? Lets take alook at the extension landscape out there.

Speaker: Caolán McNamara

Speaker: Eike Rathke

Some examples of weird behavior encountered while developing / bugfixing the LibreOffice Calc spreadsheet application.

Speaker: Matthieu Huin

This presentation shows the workflow currently followed by RDO toensure the quality of OpenStack packaging, and the specificities of thetools implementing this workflow.

Speaker: Petr Šabata, Adam Samalik

Following the progress of the Modularity initiative and Factory 2.0developments in Fedora, it’s about time we define what the next,modular operating system should look like, focusing on the developer’spoint of view.

Speaker: Peter Robinson

Generic distributions such as Fedora, CentOS, RHEL and others have welldefined update and security mechanisms as well as other processes thathave been established and proven to work over a multiple decades. Howcan we make use of these positives of distributions along with moderntools and technologies to produce a secure, stable, scaleable OS forIoT products?

Speaker: James Shubin

Next Generation Config Mgmt

A presentation about a next gen config management tool, and thespecific problems this project solves.

Three of the main design features of the tool include: * Parallelexecution * Event driven mechanism * Distributed architecture

This talk will demo a prototype I’ve built that implements these ideasand which is written in golang. I will start by presenting anintroduction to the tool. I will then demo the new features that wereadded since the project was introduced. This will include the automaticgrouping, automatic elastic etcd clustering and remote executionfeatures. I will finally share some of the future planned designs forthe tool.

An introductory blog post on the subject is available.https://ttboj.wordpress.com/2016/01/18/next-generation-configuration-mgmt/Attendees are encouraged to read it before the talk if they areinterested!

Please note that this talk was originally scheduled to be given at15:00. The talk originally in this slot, “Puppet Catalog Diffs inTheForeman” by Greg Sutcliffe will now take place at 15:00.

Speaker: Piotr Kliczewski

During this presentation we will see how to manage infrastructure whichis used to run containers. We will see how to use reliable vmsprovisioned by ovirt and run openshift containers on them by usingsingle management UI (manageiq) with ansible modules.

Speaker: Adam Litke

In oVirt datacenter virtualization environments, a manager directshosts to initiate operations to shared storage. These operations createor remove volumes, copy data between volumes, create or mergesnapshots, and various other actions related to virtual machinestorage. For efficiency and balance these operations should bedistributed across multiple hosts and run in parallel when possible.Maintaining reliability under real world conditions requires carefulmanagement and resilient algorithms. This talk will introduce some ofthe problems that can arise including: dropped communications,scheduling conflicts, and host or storage array failure. Next, asolution to these problems using shared storage locking, atomicoperations, volume generations, and forensic analysis of the storagewill be presented. Through step by step examples, the audience willunderstand how the proposed solution can solve all of the outlinedproblems.

Speaker: Roman Mohr

How do you integrate containers in your IaaS? In a VM based IaaSenvironment, introducing containers can be a painful experience. Mostlikely you end up running containers inside VMs to reuse existinginfrastructure, or you start dividing your data-center into acontainer- and a VM-world. Either way, you have two managementsolutions and non optimal resource management. But what if we put VMsinside containers? Would such a copernican revolution give us somebenefits? This talk covers our research around using Kubernetes as avirtual machines cluster manager.

Speaker: Eduardo Habkost

QEMU is an open source machine emulator and virtualizer written in C.Over time it has evolved multiple interfaces to interact with theoutside world, and multiple internal APIs and abstractions to model andkeep track of data. The talk will be a review of some of the challengesand trade-offs involved in making those abstractions work together.

Speaker: Rafael Martins

Lago is an ad-hoc virtual framework which helps you build virtualizedenvironments on your server or laptop for various use cases. It createsand orchestrates virtual machines that can be used to run test suitesand allow a developer to quickly test his code change on non-trivialflows such as live migration even before starting the CI process. Inthis session, we will walk through the usage of Lago in the oVirtproject and how it can be extended to assist virtualization developers.

Speaker: Andrea Arcangeli

Andrea will provide a high level perspective of the most notablemilestones in the long term evolution of the Linux Virtual Memory andVirtualization subsystems. In addition, Andrea will explore recentadvances in Memory Management related to the KVM VirtualizationHypervisor, such as NUMA balancing, THP, KSM and userfaultfd/postcopylive migration. Andrea will cover best practices, providing theaudience with an understanding of when and how to leverage thesefeatures in their environments.

Speaker: Martin Sivák

The workloads and scenarios for virtual machines grow more complexevery year. So do the interactions, availability, and performancerequirements. All that requires the administrators to carefully planwhere to start the VMs that depend on each other and/or specific hosts.

This talk will present the concepts that allow the administrator toexpress the rules for affinity between virtual machines and betweenvirtual machines and hosts to form complex relationships that willcover for example:

• licensing rules - limiting group of VMs to only use certain hosts
• overhead - web + database VMs running together
• performance - eg. storage VMs running on hosts with better IOperformance
• failover recovery - VMs returning to “their” hosts
• reservation - There is place for only one of us!

oVirt is an open source project for managing virtual data centers thatwill now help the administrator with exactly the above tasks. We haveintroduced the virtual machine affinity feature in the past and a hugeimprovement in that area is coming right now.

And the best part is that all this works in a fully dynamic environmentwith automatic conflict resolution and no manual management of hostpinning rules, saving the administrator his precious time.

Speaker: Stefan Hajnoczi

The introduction of non-volatile memory changes how applications,databases, and virtual machines will work in the future. NVDIMM is notsimply a faster block device. Programs can avoid block I/O entirely anduse byte-addressable NVDIMM to benefit from the performancecharacteristics of RAM. This requires new storage APIs thatapplications must use instead of traditional block I/O.

These new programs run successfully inside KVM virtual machines thanksto the vNVDIMM support already available in QEMU. Virtualization offersadditional options for managing and using NVDIMM beyond what isavailable on bare metal.

This talk covers the NVDIMM programming model and how KVM virtualmachines can use NVDIMM for faster I/O, reduced memory footprint, andfaster boot times.

Speaker: Nathaniel McCallum

This talk covers an alternative to key escrows using new cryptographictechniques implemented by the Clevis (client) and Tang (server)projects.

Speaker: Rich Bowen

Project documentation is so much more than just the formal manual. It’show you present yourself in all the forums where users may ask for help

• IRC, mailing lists, StackOverflow, and in-person events. If you wantpeople to use your project, and if you want people to join yourproject, you must be willing to listen to their needs, and be welcomingof their contributions.

Speaker: Robert Kratky

How to turn legacy docs into user-story-based, modular content tobetter serve users while reducing maintenance load and overall amountof docs.

Speaker: Nikolai Kondrashov

Now, when many IT-related services are delegated to external parties,government, medical, financial, and other organizations need tighttracking of what users and administrators do on their critical systems.Up to recording everything they see on the screen, the commands theyexecute, and files they access.

In this presentation Nikolai Kondrashov will review available solutionsfor user session recording, open-source and otherwise, their benefitsand shortcomings, and will present a new effort to create an integratedOpen-Source solution.

While there are many capable solutions for session recording, which cancentrally collect, search and playback sessions, there is no suchopen-source code. The best we have is jump servers with script(1), orsudo I/O logging, all manually set up. This presentation will show anapproach that would meet the needs of the modern enterprise.

The presentation will include a demo of a user session and accompanyingdata being recorded, stored centrally, inspected and played back.

The intended audience is developers of security, identity and policymanagement systems, as well as system administrators and securityofficers responsible for maintaining critical systems and preventinginsider attacks.

Speaker: Anton Marchukov

Repoman is a tool developed in-house and used as a core tool in oVirtCI and release processes. It aids the process of integrating RPMpackages from multiple sources into the single repo. Made to beself-contained, so it is easy to use from CI. Come and see what our usecases at oVirt are and how we use repoman to solve them. Beingdeveloped with an abstraction in mind it might be helpful to you too.

Speaker: Stephen Finucane

What does it take to implement continuous integration-style automatedtesting into a mailing list-driven software project? Not a lot,actually. In this talk, we demonstrate how a simple but easily scaledtesting system can be implemented for a such a project through thecombination of Patchwork, the web-based patch tracking system, and opensource CI tools such as Jenkins.

Speaker: Daiki Ueno

On a cloud computing environment it is often required to use a user’ssmart card on a remote server. That is, insert a smart card locally(windows or linux client), ssh to a server, and then utilize the smartcard to ‘sudo’ application or to a TLS application, or to ‘kinit’ toobtain a kerberos ticket. Other operating systems such as windowsprovide this functionality via USB-pass-through. The purpose of thistalk, is to describe where we are, and what we provide for that problem.

Speaker: Hubert Kario

The Transport Layer Protocol is becoming more and more complex. Withmore than 4 versions deployed side-by-side, the complexity of serversis increasing even faster.

To fight this, we’ve started to work on TLS test framework thathopefully we’ll later be able to turn into a dedicated TLS protocolfuzzer. Currently we have a battery of tests for obscure and not soobscure bugs as well as general RFC compliance.

Speaker: Nathaniel McCallum

Increased uses of cryptography in web environments, particularlyauthentication, have driven a set of new RFCs: JSON Object Signing andEncryption. While these standard data formats have dominated web-basedapplications, they also have further applicability in a wide variety ofnon-web contexts. This talk will introduce you to José, a plug-able,open source C implementation of these RFCs which provides both an APIfor direct integration and a usable command-line interface. We willdiscuss the techniques that we used to keep our API usable across allthe possible input parameters and show example of how you can implementJosé in your own infrastructure. We will also discuss some of our plansfor the future, including how you can help contribute.

Speaker: Anton Marchukov

The world is not perfect and network failures do happen. Complainingabout instability might not always be enough. Especially when yoursystem depends on multiple networked services and each of them is on acritical path to the final result.

This talk will follow a real story of an attempt to implement networkerrors handling by retries functionality that is inside urllib3 andrequests Python libraries. More importantly we will simulate poornetwork conditions on a local machine using Linux Network Emulator andthen will reason on how effective the attempt was and what can befurther improved.

This is a proper treatment the networked systems should get when theyare designed, developed and tested. And it is great that all the toolsnecessary are already there in most popular Linux distributions.

Speaker: Thiago Santos

This talk will explore how Mapbox GL Native, a hardware-accelerated maprendering engine, can bring beautiful maps made with open data to theopen source community. We’ll also cover the possibility of bringingMapbox GL Native to other open source platforms, such as GTK. Lastly,we’ll dive into our recently collaboration with the Qt Company to bringthe power and flexibility of Qt to Mapbox GL Native.

Speaker: Andreas Nilsson

An important aspect of humane interfaces is meeting the needs of avariety of people. They all have different skills, restrictions andwhims. How do you figure out what exactly those are?

In this presentation I’ll talk about how I used user interviews andpersonas for that purpose, using them as tools for discussions andimplementation of Public Transportation in GNOME Maps. I’ll talk abouthow I interviewed people from both cities and countryside, from bothBrazil and Sweden, and how their insights challanged my own perceptionsabout what people need from a public transportation UI.

Speaker: Tomer Brisker

You think you hit a bug in open source project. Now what? In this talkwe will go over everything from where to get support when you hit anissue, through submitting a useful bug report, to how to contribute afix that will get accepted quickly. I will also talk a bit about mywork as a maintainer.

Speaker: Brian Proffitt

In this presentation, Brian Proffitt will explain best practices forbeing a good community mentor, setting up scope-appropriate projects,and troubleshooting when things start going off the plan.

Speaker: Dave Neary

We are all a product of our experiences. Different communities aroundthe world have different core assumptions about behaviour, howdecisions are made, the role of the individual in a group, and more.What makes up culture, and can we have better community experiences byunderstanding it?

Speaker: Dodji Seketeli

Speaker: Orit Wasserman

Ceph is a highly available distributed software defined storage,providing object, key/value and file-system interfaces. Ceph RadosGateway (Radosgw) provides HTTP REST API that is S3 and openstack swiftcompatible. This talk will cover cloud object storage concepts and howCeph implementation of cloud object storage (Radosgw). This talk willalso present the newest features and our plans for the future.

Speaker: Jose Rivera, Mohamed Ashiq

While containers themselves are stateless many applications still haverequirements on storage that should persist across containers andinstances of containers. Many such storage solutions require anadministrator to set up a storage solution on hardware outside theirexisting container platforms. GlusterFS changes all that.

Speaker: Orit Wasserman

Ceph is a highly available distributed software defined storage,providing object, key/value and file-system interfaces. Ceph RadosGateway (Radosgw) provides HTTP REST API that is S3 and openstack swiftcompatible. This talk will give a brief Radosgw architecture overview,present the newest features and our plans for the future.

Speaker: Josh Berkus

Abstract: So you’ve containerized your application, and now you want todeploy it scalably across a cluster. Maybe you’ve looked at Kubernetesbut you can’t figure out how to use it. In one short session, we’llteach you enough to get started.

Speaker: Ratnadeep Debnath

Containers are great in terms of application packaging and delivery,but there’s a lot of noise in the space. But when it comes tomulti-container applications, most production setups use advancedcontainer orchestration technologies like Kubernetes, Openshift,Mesos/Marathon, which are not that developer friendly.

Developers prefer docker-compose for its simplicity. This talk willshowcase our ongoing efforts at Red Hat, Skippbox and Google to bridgethis gap between deploying containers in development to production, andthe need to standardize a multiple container definition spec whichworks seamlessly across different environments and containerorchestration platforms.

Speaker: Lalatendu Mohanty

OpenShift version 3 is an open-source Platform-as-a-Service (PaaS)based on docker-formatted container images and the Kubernetes project.Kubernetes is an open-source project for orchestration, deploymentautomation, auto-scaling, and management of containerized applications.OpenShift combines all of the benefits of docker and Kubernetes andadds features like automatic image building, a local docker registry,storage integration, networking, and Continous Integration (CI) toolssuch as Jenkins, to deliver a platform for Continuous Delivery (CD).Simply put, just bring your application code and OpenShift will buildthe required Docker images and run the application on OpenShift withKubernetes. OpenShift is also ideal for deploying the microservicesthat containers make so easy to implement.

Speaker: Bilgin Ibryam

Kubernetes is awesome! But what does it takes for a Java developer todesign, implement and run Cloud Native applications? In this session,we will look at Kubernetes from a user point of view and demonstratehow to consume it effectively. We will discover which concernsKubernetes addresses and how it helps to develop highly scalable andresilient Java applications.

Speaker: Julien Danjou

Gnocchi is a time series database written in Python, that has beencreated in the context of the OpenStack cloud computing project. Itoffers highly-scalable data storage for measurements and providesaccess to its data via a REST API. In this lecture, we’ll discuss thefeatures the project is offering to its users, and how they can easilybe leveraged in any application. In a second part, we’ll see how theproject has been built to scale, how Python was leveraged and madescalable.

Speaker: Brian Bouterse

When things go wrong in production, it can be necessary to troubleshootproblems where they occur, instead of in a development environment. Inthose situations having a working knowledge of GDB, GDB PythonExtensions, and strace is very helpful. You will see some simpletechniques to get insight into those situations. This talk outlinesseveral techniques for connecting to an already running, “stuck”, ordeadlocked Python process using GDB for debugging.

During the talk, we will:

• inspect the current state of threads with
• use and demo the GDB macros for Python
• inspect a locally running process and a core dump collected from aremote machine
• use strace to gather system call information about a process
• discuss the SIGTRAP handler as a proactive way to make rpdb availablein production.

I have had to debug several hard-to-find bugs that were very infrequentdeadlocks using Python. Furthermore it was happening on remote machinesI could not have network access to. This technique was invaluable inthose situations.

Everything is in the abstract

Speaker: Victor Stinner

Working on optimizations is a task more complex than expected on thefirst look. Any optimization must be measured to make sure that, inpractice, it speeds up the application task. Problem: it is very hardto obtain stable benchmark results.

The stability of a benchmark (performance measurement) is essential tobe able to compare two versions of the code and compute the difference(faster or slower?). An unstable benchmark is useless, and is a risk ofgiving a false result when comparing performance which could lead tobad decisions.

I’m gonna show you the Python project “perf” which helps to launchbenchmarks, but also to analyze them: compute the mean and the standarddeviation on multiple runs, render an histogram to visualize theprobability curve, compare between multiple results, run again abenchmark to collect more samples, etc.

The use case is to measure small isolated optimizations on CPython andmake sure that they don’t introduce performance regression in term ofperformance.

Speaker: Ilya Etingof

The ASN.1technology will be briefly explained followed by the introduction tothe pyasn1 package, its typicaluse cases and key features.

Speaker: Freddy Rolland

The oVirt Project is an open virtualization project providing a feature-richserver and desktop virtualization management platform with advanced capabilitiesfor hosts and guests, including high availability, live migration, storagemanagement, system scheduler, and more.

oVirt provides an integration point for several open source virtualizationtechnologies, including kvm, libvirt, spice and oVirt node.

The session will provide an intro to the project components and features.