k8gb: Global Load Balancing for Kubernetes
Shriira Press
How a small Kubernetes operator turns the oldest protocol on the internet into a global load balancer that survives whole-region outages.
Welcome to k8gb: Global Load Balancing for Kubernetes.
When a single Kubernetes cluster is no longer enough — because you run in several regions, because one data centre going dark must not take your application with it, or because users on two continents deserve a fast local entry point — you need something that steers traffic across clusters that may be thousands of kilometres apart. k8gb is the CNCF sandbox project that solves this without proprietary appliances or a central control plane. It leans on DNS, the time-proven naming protocol that already spans the planet, and adds a Kubernetes operator, an embedded CoreDNS with a custom plugin, and a single Gslb custom resource. This book walks through the whole picture: the global load balancing problem and why DNS is the right tool, the architecture that lets every cluster act independently with no single point of failure, the Gslb CRD and its round-robin, failover, weighted, and geoip strategies, the zone-delegation mechanics that knit clusters together through your edge DNS provider, how health checks ride on ordinary liveness and readiness probes, and finally how to run k8gb sensibly in production. By the end you will understand not just how to install it, but why each piece is shaped the way it is.
This title is part of the ShriIra library and is free to read in full, right here — our small contribution to making world-class knowledge easy to reach.
A note on reading it: open the Contents menu at the top of the reader to jump between chapters, use the Aa menu to set a comfortable text size, theme (light, sepia, or night), and single- or two-page layout. Your place is saved automatically, so you can always pick up where you left off.
We hope it serves you well.
— Shriira Press