Automating Day 2 Operations with Kubebuilder and Operator SDK
E142 | Sun 16 Jul 5:30 p.m.–6 p.m.
Presented by
-
Steve Sklar
https://sklar.rocks
Coming from a background in finance and economics, Steve taught himself how to code on a bond trading desk and hasn't looked back since! He's architected, developed, deployed, and maintained critical software systems for financial firms, and ML/AI startups. Currently, Steve works for QuestDB, an open source time-series database company with a focus on high performance and usability. His hobbies include playing music, learning how to golf, and spending time with his dog, Nacho.
Steve Sklar
https://sklar.rocks
Abstract
We all know how powerful Kubernetes is when it comes to orchestrating cloud infrastructure. So how can we use it to automate higher-level tasks, like upgrades, snapshots, or data migrations? Using tools like Kubebuilder and Operator SDK, we can leverage core k8s building blocks to safely automate these "Day 2 Operations."
In this talk, I will first provide a brief overview of operator mechanics before taking a deep dive into a sample use-case: managing database snapshots and restores. I will explore key design decisions behind the API objects that are used to model the problem, followed by a discussion of how Kubebuilder and Operator SDK can help us write an operator to manage the complete snapshot lifecycle using familiar tools like yaml and kubectl.
By the end of the talk, the audience should feel more comfortable around custom operators, and may even be inspired to automate some their own operations!
We all know how powerful Kubernetes is when it comes to orchestrating cloud infrastructure. So how can we use it to automate higher-level tasks, like upgrades, snapshots, or data migrations? Using tools like Kubebuilder and Operator SDK, we can leverage core k8s building blocks to safely automate these "Day 2 Operations." In this talk, I will first provide a brief overview of operator mechanics before taking a deep dive into a sample use-case: managing database snapshots and restores. I will explore key design decisions behind the API objects that are used to model the problem, followed by a discussion of how Kubebuilder and Operator SDK can help us write an operator to manage the complete snapshot lifecycle using familiar tools like yaml and kubectl. By the end of the talk, the audience should feel more comfortable around custom operators, and may even be inspired to automate some their own operations!