Kubernetes Setup Guide

This page is a stub — help us fill it in

This page has been identified as a community content gap. Use the Edit this page link at the bottom to add your knowledge and claim the bounty.

This page will guide you through running Kubernetes — from a single-node k3s install on a Raspberry Pi or small VPS through a multi-node cluster for real self-hosted workloads. It will cover the minimum viable Kubernetes knowledge to get applications running, persistent storage, ingress, and know when Kubernetes is the right tool vs when docker-compose will serve you better.

Topics to cover

• k3s single-node install — why k3s for self-hosting (lightweight, batteries-included), curl-install command, what Traefik and CoreDNS you get by default, kubeconfig setup
• kubeadm cluster bootstrapping — when to use kubeadm vs k3s, control plane setup, worker node join, CNI selection (Flannel, Calico, Cilium)
• kubectl basics — essential commands (get, describe, logs, exec, apply, delete), context management, useful aliases and shortcuts
• Namespaces and RBAC — why namespace separation matters, ServiceAccount basics, ClusterRole vs Role, common RBAC patterns for multi-tenant clusters
• Helm chart deployment — what Helm is, adding repos, installing and upgrading charts, values.yaml customization, helm diff for preview
• Persistent storage — local-path provisioner (k3s default), Longhorn for replicated storage on multi-node clusters, PersistentVolumeClaim basics
• Ingress controllers — Traefik (k3s default), Nginx ingress controller, cert-manager for TLS: how they work together, example Ingress resource
• When to use K8s vs docker-compose — complexity trade-offs, operational overhead, good candidates for K8s (HA, many services, team usage) vs docker-compose (personal, few services, simplicity priority)