On today’s episode of the Kubelist Podcast, I had a conversation with Lee Mills and Matt Clarke from Spotify. Lee is the engineer manager working on the Backstage project, while Matt is a Sr Infrastructure Engineer. Backstage was recently donated to the CNCF as a Sandbox project. On the podcast, Lee and Matt explain the inspiration for Backstage and really help us understand the origins and direction of the project. Spoiler: it’s a pretty good idea and they’ve done a great job getting it off the ground. To keep the theme going this week, here are some of our favorite newer and older links about the Backstage project!
The Backstage team is answering questions today over at r/kubernetes. This is a great opportunity to chat with the engineering team, outside of the normal community meetings. After spending time listening to Lee and Matt talk about Backstage on the podcast, I’m excited to see the questions asked today!
Matt spoke about this new plugin on the podcast episode. Backstage is a developer tool, and developers own their service in production, right? So how do you expect your developers to own a service spread across multiple clusters, running in multiple regions or kube contexts? The answer: this plugin for Backstage, to help manage these multi-cluster service deployments – thinking about how the service owner wants to see it, not how the cluster operator looks at the infrastructure. This might be an easy and quick win to adopt Backstage on a team. 🔌
A good introduction to Backstage from when it was released. The project has matured considerably since this post (it’s almost a year old, things are moving quickly). This post clearly explains why Spotify created and open-sourced the Backstage project. As mentioned in the post: “it may seem odd that a music company is launching a best-in-class developer portal.“ What’s great though is that after this post, the project moved from OSS to being a CNCF Sandbox project. 🎸
This is a short post from a startup that’s building on the Backstage ecosystem, and contributing back. There’s a lot of open source out there. If you’re looking to figure out how to get adoption or some early traction in your project, this is a good read. The roadmap and path outlined that worked for Backstage isn’t likely going to map to your project, but retrospective stories like this are great to learn from and see if there’s a way to apply them to the next project. 🗺
The title describes it well: this is an early post that describes how Spotify uses Backstage internally. Their use of the project has definitely expanded, as explained by Lee and Matt in this week’s podcast. But the core functionality was there on day one, and the project continues to add functionality in new areas. This is a great early look at the project use, from before it was a CNCF project. 🎶
This isn’t directly related to the Backstage project, but Lee Mills (guest on this week’s podcast episode) was on a Kubecon End User panel discussion. The question linked here is asking which cloud-native principles were the most beneficial, and which caused the most concern in the org. Lee starts out with a broad response about thinking about automation. Lee is now the engineering manager working on Backstage, a project that’s really all about automation at a different level than we often think about. ⚡
The KubeCon EU (Virtual) schedule is going to be announced today, and the CFP for KubeCon NA (maybe in person?) has opened!