Les sujets de cette chaîne sont associés aux tags suivants :
Pourquoi et comment passer au PaaS, les bonnes pratiques
Sujets de type : « Présentation Produit »
How To Develop a Jenkins Plugin
Par Grégory Boissinot, Zenika [en français]
Today Starts a New Era for Companies Who Want to Setup a Global Continuous Delivery Practice.
In the last few years, CloudBees has witnessed first hand the evolution and adoption of DevOps and continuous delivery (CD) in organizations.
Originally, most of our discussions were “Jenkins” discussions. Teams within organizations had made the decision to use Jenkins as their de facto tool for continuous integration (CI) and/or continuous delivery (CD). As Jenkins became their unique gateway to production (i.e. anything that lands in production has to travel through a Jenkins pipeline to get there), Jenkins became as critical as production itself for them: if you can’t upgrade or fix anything into production, you have a big (production) problem! To make those teams successful, we provided a number of extensions on top of Jenkins (such as role-based access control and other features), as well as 24/7 support backed by our worldwide team of Jenkins experts. This is today an extraordinarily successful CloudBees offering that helps hundreds of teams and thousands of users around the globe operate a rock-solid Jenkins cluster.
In the last few years, however, the tone of these discussions has changed. We are now meeting with a lot of enterprises that are looking at building a formal continuous delivery “practice” in their organization. They want to standardize the way continuous delivery happens, across the board. They want to be able to compare the productivity and velocity of all of their teams. For them, gone are the days of team-specific continuous delivery solutions. They have learned a lot through what leading-edge teams have done, they have setup proof of concepts and they are now ready to leverage their critical mass to formalize, at scale, the best practices that fit their business.
What they are looking for is a single, unified continuous delivery solution that gives them visibility into all of their teams and applications, this, in turn, requires a platform that knows how to integrate with legacy, traditional and leading edge environments, from AIX to Docker on AWS - not one different CD solution per project or technology of the day! If speed and agility matter for individual applications, it certainly matters to the IT organization itself! As such, they can’t afford to have a one month lag time anytime they onboard a new team. They can’t even afford one day. They want to onboard new teams or new projects in a snap and give them a best of breed environment to build those delivery pipelines. Also, they want a platform that’s cost efficient. Efficient, both in terms of how it manages the underlying infrastructure at scale, and also in how much (or, rather little) work is involved in managing the platform itself.
Consequently, in order to fulfill that need, CloudBees is launching today CloudBees Jenkins Enterprise, the first and only platform that enables continuous delivery at scale for enterprises, based on the de facto DevOps hub, Jenkins.
CloudBees Jenkins Enterprise is a full-fledged platform that can be deployed anywhere: Linux, VMware, OpenStack, AWS, to name a few. It takes ownership of the provided infrastructure and provides a fully-managed Continuous Environment built on Jenkins. Based on Docker containers, CloudBees Jenkins Enterprise provides a self-service, elastic CD environment that can be centrally managed. It also enables enterprises to set up global policies and best practices that can be enforced among all teams across the organization. Furthermore, the platform automatically handles backup and restore, automatically detects faulty behaviors and properly recovers from those situations. This leads to a continuous delivery platform with a very low cost of maintenance and an excellent usage of the infrastructure through high-density Jenkins deployments that can readily scale up to thousands of teams, and tens of thousands of projects and users.
If you are interested to know more, I’d suggest reading the excellent blog post by Brian Dawson, Product Marketing Manager at CloudBees.
This week, we launched CloudBees Jenkins Enterprise to enable enterprise-wide DevOps through CD as a Service.
Why is this important to you? Simply put, the results are in. Organizations which implement continuous delivery (CD) in support of enterprise-wide DevOps see significant improvement in release frequency, cycle time and mean time to recovery. More importantly, such improvements lead to a more agile, more responsive more competitive overall business.
To successfully implement continuous delivery and DevOps in a large, mature enterprise, there are specific needs and obstacles which must be addressed. Let’s look at them:
Traditionally, meeting the requirement for scalability has been the biggest challenge. Much of this has to do with the way continuous delivery has been adopted and the nature of the available CD solutions.
CD and DevOps adoption has often begun within individual teams as grassroots efforts. The tools used for such grassroots implementations fall largely into two categories:
These solutions present issues when growing CD from one team to an entire organization. Common problems are:
CloudBees Jenkins Enterprise enables you to scale without instability by implementing the only solution with a Distributed Pipeline Architecture (DPA). To better understand DPA, it helps to look at what happens when traditional solutions scale.
When we setup CD for a single team, things look good. We can deliver a single service through our CD pipeline with speed:
But as we add teams, instability of our CD server increases. Our speed decreases. We are unable to update business-critical services. Single server, single point of failure.
The elasticity of the Distributed Pipeline Architecture distributes teams’ CD workloads across multiple isolated servers, providing high levels of scalability. Now multiple teams using multiple pipelines can deliver multiple business-critical services reliably. Scaling with DPA enables speed AND stability.
Building on the scalability enabled by DPA, CloudBees Jenkins Enterprise supports enterprise-wide DevOps with other best-in-class features:
The launch of CloudBees Jenkins Enterprise is important to you because enterprise DevOps built on the practice of continuous delivery is how you remain competitive in today’s market. To do this you require the scalability, security, manageability and resiliency provided by Enterprise and the only Distributed Pipeline Architecture. Deploy CD as a service in minutes on your existing infrastructure.
In every Meet the Bees blog post, you’ll learn more about a different CloudBees Bee. This time we are in France, visiting Arnaud Héritier.
Who are you? What is your role at CloudBees?
My name is Arnaud Héritier and I’m a Support Delivery Manager in the Customer Engagement team at CloudBees.
I have used various open-source projects and contributed to them since the beginning of my career. My main contributions are to Apache Maven and Jenkins, thus it was obvious for me to join CloudBees when the opportunity came in May 2015.
Previously I had different roles (Developer, IT Architect, IT Consultant, Professional Services, Release Manager, Forge Manager, …) in various types of companies (software vendors, IT consulting, …). I also worked (in)directly for various companies in non-IT sectors like media, banking, insurance, …
Developer support engineers are in front of our customers to help them in their daily tasks.
This is a really interesting position for various reasons:
What does a typical day look like for you?
Our support delivery team is based in 3 timezones (Europe, US east and Australia) to provide 24/7 support. Each team starts its day with a short meeting to synchronize on open/new cases and to request some help. This is a great opportunity to have an overview of what is in progress between 2 regions and to share our knowledge on the different kinds of cases we are managing.
While one of us is taking care to review all incoming cases all others are processing them. We have many different kind of activities:
What do you think the future holds for Jenkins?
A really bright future!! I have been a part of the Jenkins community for a very long time (I think I still had some hair when I talked for the first time with Kohsuke on the mailing list) and Jenkins is the only CI/CD tool with such a vibrant community. All of this is due to its extensibility which allowed to create a really large ecosystem, and Kohsuke’s kind leadership. Today, automation tools are evolving to cover more modern usages like continuous delivery/deployment and Jenkins is leading this direction. When you remember that 3 years ago, Pipeline didn’t exist and when you see what Jenkins users are achieving with it today, that’s really impressive! When you see how Blue Ocean is now providing a new user experience … that’s just awesome!
What are some of your best tips and tricks for using Jenkins?
Do you have any advice for someone starting a career in the CI/CD/DevOps/Jenkins space?
Collaborate, communicate and automate!
This is the heart of DevOps and while Jenkins will help you to automate various processes to achieve your CI/CD objectives, the most important thing will be: you!
What are some of the most common mistakes to avoid while using Jenkins?
To believe that Jenkins will do auto-magically everything for you. Jenkins is a framework, providing many services, but you are responsible for making it useful!
What is your favorite form of social media and why?
Twitter without a doubt. You have the opportunity to talk with so many different people. Some known, some unknown, but you can exchange on many subjects, technical or not, serious or not, … I really love this.
But let’s be honest nothing replaces meeting people IRL and that’s why I love to attend various conferences as an attendee but also as speaker to share my passion.
In this episode, Brian Mericle, distinguished engineer at Choice Hotels International, “checks in” with DevOps Radio host, CloudBees CMO and occasional Jenkins butler, Andre Pino to provide the inn-side scoop on what Choice Hotels - the 75+ year-old franchise that includes Comfort Suites, Comfort Inn, Sleep Inn and Cambria brands - is doing in terms of deployment. He’ll also share his secrets for managing big DevOps teams, without reservations.
Brian explains how Choice Hotel’s DevOps adoption was driven by the need to maintain a competitive advantage in terms of its website and services. Choice Hotels provides web-based services to its franchisees and, in this case, needed to update the central reservation system used by the various Choice Hotels properties to book rooms. The goal was to get to market faster and be a driving force in the market, but manual patterns and processes were presenting roadblocks. By implementing DevOps, Choice Hotels was able to accelerate the upgrade of its central reservation system.
Choice Hotels employees a 250+ person development staff. Brian admits that with a team that big, transitioning to a DevOps culture does not happen overnight. Brian believes there is no such thing as over-communication, and that everyone in IT should always know what’s happening. He tries to meet with all his teams in order to ensure transparency. Sounds like five-star hospitality to us!
This “suite” episode is available now on the CloudBees website and on iTunes. You can also join the conversation on Twitter by tweeting out to @CloudBees and including #DevOpsRadio in your post. Before you check-out make sure you subscribe to DevOps Radio where you can catch up on past episodes.
I am very excited to announce the addition of Declarative Pipeline syntax 1.0 to Jenkins Pipeline. We think this new syntax will enable everyone involved in DevOps, regardless of expertise, to participate in the continuous delivery process. Whether creating, editing or reviewing a pipeline, having a straightforward structure helps to understand and predict the flow of the pipeline and provides a common foundation across all pipelines.
Pipeline as Code was one of the pillars of the Jenkins 2.0 release and an essential part of implementing continuous delivery. Defining all of the stages of an application’s CD pipeline within a “Jenkinsfile” and treating it as part of the application code automatically provides all of the benefits inherent in SCM:
We recommend people begin using it for all their pipeline definitions in Jenkins and the CloudBees Jenkins Platform. The plugin has been available for use and testing starting with the 0.1 release that was debuted at Jenkins World in September, it has already been installed over 5,000 times.
If you haven’t tried Pipeline or have considered Pipeline in the past we believe this new syntax is much more approachable with an easy adoption curve to quickly realize all of the benefits of Pipeline as Code. In addition, the predefined structure of Declarative makes it possible to create and edit pipelines with a graphical user interface (GUI). The Blue Ocean team is actively working on a Pipeline Editor that will be included in an upcoming release.
If you have already begun using Pipelines in Jenkins we believe that this new alternative syntax can help expand that usage. The original syntax for defining pipelines in Jenkins is a Groovy DSL that allows most of the features of full imperative programming. This syntax is still fully supported and is now referred to as “Scripted Pipeline Syntax” to distinguish it from “Declarative Pipeline Syntax.” Both use the same underlying execution engine in Jenkins and both will generate the same results in Pipeline Stage View or Blue Ocean visualization. All existing pipeline steps, global variables and shared libraries can be used in either. You can now create more cookie-cutter pipelines and extend the power of Pipeline to all users regardless of Groovy expertise.
Other key features of Declarative Pipeline include:
Be on the lookout for future blog posts here or on Jekins.io detailing specific examples of scenarios or features in Declarative Pipeline. Andrew Bayer, one of the primary engineers behind Declarative Pipeline, will be presenting at FOSDEM in Brussels, Belgium this weekend. We have also scheduled an online Jenkins Area Meetup (JAM) later this month to demo the features of Declarative Pipeline and give a sneak peek at the upcoming Blue Ocean Pipeline Editor.
In the meantime, we have updated all Pipeline documentation to incorporate a Getting Started guide, a Guided Tour and a Syntax Reference page with numerous examples to help you get on your way. We have also created a Quick Reference card that can be printed and hung nearby. Simply upgrade to the latest version of the Pipeline plugin in Jenkins to enable all of these great features.
Last fall, prior to Jenkins World, CloudBees conducted a community survey on behalf of the Jenkins project. We were grateful to receive over 1,200 responses – and thanks to this input, we gained some interesting insights into what Jenkins users are doing.
Based on the results, it’s safe to say that Jenkins is currently viewed as the #1, continuous integration (CI) server and is rapidly becoming the leading continuous delivery (CD) tool. Adoption of Jenkins 2, which introduced CD pipelines, clear visibility of delivery stages and multiple usability enhancements has skyrocketed to nearly half the active user base. Once again, there was a lot of consistency in many findings from year-to-year. For example, the number of Jenkins users continues to increase, with 90% of survey respondents considering Jenkins mission-critical.
Here are some of the key findings:
We want to thank everyone for completing the survey - and congrats to Iker Garcia for winning a free pass to Jenkins World 2017 and to Dave Leifer for winning the Amazon gift card.
See you at Jenkins World, August 28-31, in San Francisco, California! Register now for the largest Jenkins event on the planet in 2017 and get the Early Bird discount. The Call for Papers is open until March 5 – so submit a talk and share your Jenkins knowledge with the community.
Jayne Groll (below) is CEO and co-founder of DevOps Institute, and an expert in IT training. She joins DevOps host, Andre Pino, to discuss how people can get training and certification in DevOps topics - and what exactly that means in an industry where there isn’t a single body of knowledge about DevOps.
Jane explains how a long time ago in a galaxy far away, when screens were still black and green, she was a paralegal manager tapped to provide education and support her company’s IT unit. Since then she has witnessed IT education and certification grow with new roles, competencies and skills, including DevOps.
Jane says for most developers the most difficult part of DevOps is starting the process. The DevOps Institute’s curriculum encompasses a collective body of knowledge including CD and CI, automation, principles of how to architect a DevOps pipeline, test communities, cultural management, DevSecOps (adding security to DevOps) and specific role-based competencies. By participating in this training and certification process, developers and operations teams can gain the skills needed for successful DevOps strategies, so that they can ultimately bring value back to the organization.
Finally, Jayne talks about DevOps Express – a consortium of leading vendors and practitioners in the DevOps space who are providing, collectively, best practices and a reference architecture for organizations trying to undergo a DevOps transformation. DevOps Institute is a founding member of DevOps Express, and Jayne joins with Andre in discussing the value of it for the DevOps community. Listen in to hear more!
Here’s your homework assignment: Catch up on DevOps Radio, available on the CloudBees website and on iTunes, and make note to check back for new episodes!
Want extra credit? Overachievers, make sure you follow @ITSM_Jayne and @CloudBees on Twitter, so you can share your thoughts on the latest episode. All you have to do is mention #DevOpsRadio in your post.
We are proud to announce the immediate availability of CloudBees Jenkins Platform 2.32.1, which offers upgrade notifications and many key improvements such as a bump on the Jenkins core to the 2.32.1 LTS line. You may know that Beekeeper Upgrade Assistant allows users to review and install upgrades of verified components, tested through the CloudBees Assurance Program (CAP). Up to now, this recommended configuration or envelope was only updated when new CloudBees Jenkins Platform releases were announced. Starting with this release, CloudBees can update the configuration between releases, so that fixes can be safely deployed, keeping the Jenkins instance safely in the recommended configuration. Another key improvement, is the LTS upgrade on the rolling release. This LTS bump provides the latest and greatest stable Jenkins core, fixes 50+ bug fixes since the previous LTS release, includes some performance improvements and, most importantly, upgrades the Jenkins Remoting Module from version 2.6 to 3.1. The Jenkins Remoting model is at the heart of all communications not only between Jenkins masters and agents but also between CloudBees Jenkins Operation Center and each client master.
To learn more, read the full blog post on Cloudbees Network (CBN). As the hub of all product related knowledge, CBN is where you will find release details and all future product announcements.
Wen Gu is the Software Engineering Manager of Engineering Effectiveness at Twitter and keynote speaker at Jenkins World 2016. He brings his experience in software development for tech giants to DevOps Radio’s first episode of 2017.
In this episode, special host Sacha Labourey, CEO of CloudBees, chats with Wen about his moves from DevTools to DevOps, Intuit to Twitter and Hudson to Jenkins, covering the evolution of CI/CD and the tricky definition of DevOps. Wen also describes the DevOps cultural experience at legacy tech companies like HP, Yahoo and Intuit, as well as at younger companies like Twitter. He stresses the importance of unifying DevOps’ tools and best practices to get to the crucial last mile of the software delivery process – deployment.
New to DevOps Radio in 2017? Catch up on last year’s episodes, available on the CloudBees website and on iTunes, and make sure to stay tuned for new episodes.
We’d love to hear your thoughts on the latest episode of DevOps Radio. Given Wen’s place of employment, it seems only appropriate to join the conversation on Twitter, which you can do so by including the @CloudBees handle and #DevOpsRadio in your post.
Sacha Labourey, CloudBees, and Wen Gu, Twitter, going the last mile together, on DevOps Radio
All of us here at CloudBees are committed to making Jenkins the de-facto tool for continuous integration (CI) and continuous delivery (CD). In addition to our CloudBees Jenkins Platform and Private SaaS Edition products we continue to invest heavily in Jenkins through contributions to Jenkins core, Pipeline, Blue Ocean and many vital plugins for Jenkins. Our success is dependent on Jenkins’ success.
This investment in open source software (OSS) is manifest in an engineering team and resources dedicated to contributing bug fixes, security fixes and new features to Jenkins and its ecosystem. All of the work done by this team goes directly towards improving Jenkins and we want to keep everyone up-to-date with the changes we are making. To that end we have begun hosting monthly demonstrations to show off the latest and greatest changes we are working on and answer any questions.
We hosted the first of these “Demo Days” at the end of October and the second at the beginning of December. These videos are publicly available and viewable at anytime. If you have questions, you can always ask someone in IRC #jenkins.
Going forward we are going to host the live broadcast for these demonstrations regularly on the second Thursday of each month at 8 AM PST/4 PM GMT. We are also moving to CloudBeesTV as the host and archive for the videos. I have created a new Playlist for “CloudBees OSS Demonstrations” to allow you to quickly find all related videos.
The first OSS Demo Day on CloudBeesTV will be this Thursday, January 12, 2017. We hope you’ll join us and ask questions.
Sujets à venir, dont la date a été planifiéeAucun pour le moment
Sujets en attente de financement dans cette chaîneAucun pour le moment
Les sujets dont la date est passée, en préparationAucune pour l'instant
Sujets publiés et livrés dans cette chaîne (12)
Retrouvez ici les webinars de cette journée organisée par CloudBees
By Jérôme Lacoste, We Want To Know
By Julien Carsique, Nuxeo
By Lars Kruse, Praqma A/S
By Xavier Seignard, Pod Programming & Mickael Istria, JBoss by Red Hat Tycho, Jenkins & Sonar
Benoit Moussaud, Xebia Labs [en français]
By Henri Gomez, Axway
Sebastian Bergmann, thePHP.cc
Par Bruno Guedes, Sfeir
By Arnaud Heritier, eXo Platform & Mathieu Hausherr, OCTO
Par Vincent Hardion, Synchrotron Soleil [en français]
Par Grégory Boissinot, Zenika [en français]