Introducing Josh Wood: OpenShift and Kubernetes Developer Advocate
March 29, 2018 | by
Like so many people, Red Hat was the first Linux I saw and the first Linux I installed on my own hardware. Ever since then, as several friends donned red fedoras, Red Hat sounded like a great place to work, and in the meantime the company became the gold standard for succeeding because of open source values. In other words, the chance to join Red Hat represented one of my dream jobs. Still, I was passionate about my Kubernetes work at my previous employer, and I knew I’d miss the experience and many of my colleagues there. Fortunately, I didn’t have to miss them for long.
I came aboard at an exciting moment for the OpenShift team, just as Red Hat’s acquisition of CoreOS was made official. As a Developer Advocate for OpenShift, I hear questions and feedback directly from users, and CoreOS technology answers a number of their concerns and wishlist items. The engineering talent CoreOS adds is obvious from the company’s consistent and central contributions in the Kubernetes space. You could say that Kubernetes begins with etcd, the distributed key-value store that originated at CoreOS, and CoreOS helped spearhead, along with Red Hat and others, efforts to standardize and modularize key Kubernetes interfaces such as the container runtime (CRI) and container network (CNI).
Of course, I may be biased, since I was part of CoreOS from 2015 until becoming a Red Hatter. I helped document CoreOS projects, products, and had the opportunity to speak to lots of groups about CoreOS’s leitmotif automatic updates and the open source projects and patterns that make that mantra work. With the acquisition, there is an obvious segué from those talks and training sessions on behalf of CoreOS to the OpenShift demos, documentation, and workshops I’m helping produce at Red Hat.
OpenShift delivers a PaaS layer atop its Kubernetes core, a developer- and application-centric set of features that scratches enterprise teams’ itch for source-to-deployment velocity. In Tectonic, also built around Kubernetes, CoreOS often focused on the concerns of constructing and maintaining enterprise clusters. Tectonic, famously, enables Kubernetes clusters to automatically update themselves over the air, like your mobile phone or web browser. It’s exciting to see the roadmap come together combining the best qualities of both companies and these two essential products. And, on a personal level, I worry a lot less now about slipping into discussions of my favorite CoreOS technologies.
Look for me and my colleagues from the OpenShift advocacy team at an event near you soon. The CoreOS acquisition is just one of many landmark developments in the vibrant Kubernetes community, so I’m sure we’ll have plenty to chat about.
In this blog post I want to introduce a nice little trick to make canary deployments possible only with Kubernetes StatefulSets and Services. First, let me give a bit of background on why I went down ...