Is Kubernetes sufficient by itself? This post highlights the features that a DevOps team typically needs to add on top of a vanilla Kubernetes cluster to obtain a secure, production-ready Kubernetes cluster.
Let us dive deeper into observability: What it is, the different types and what it technically means to implement observability.
HOWTO set up Kubernetes using kubespray and Terraform on clouds without a managed Kubernetes service. Uses Rook for Persistent Volumes.
Learn how to set up Rook with Ceph on Kubernetes to get dynamic Persistent Volume support even in cloud environments without a block storage service or in bare metal, on-premise setups.
Explains how Virtual Machine (VM) images are booted onto virtual disks including details about disk layouts and the cloud-init boot process.
TL;DR: EKS Distro = kops + Kubernetes fixes + aws-iam-authenticator AWS has recently launched EKS Distro, the Kubernetes distribution that powers its managed Kubernetes offer, called Elastic Kubernetes Service (EKS). This means that customers no longer have to choose between EKS (as a managed service) and some other Kubernetes setup. Instead, they can choose between EKS as a service managed by AWS or managed by themselves. The application hosted on top won’t observe any difference.
This post will detail an approach for building a base image on top of Ubuntu to be used as a start image for provisioning Kubernetes nodes (masters or workers) running as virtual machines. We will be using packer to automate the build process and to ensure basic compliance testing with InSpec tests. The target audience are people who are/will be running non-managed Kubernetes clusters or people that are looking into how to leverage Packer as a tool for automated, reproducible and tested virtual machine images.