Keyboard shortcuts

Press ← or → to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

Kubernetes Basics

Introduction and Core Concepts

Three Big Ideas

  1. Kubernetes relies on Controllers
  2. Kubernetes is a container orchestration engine
  3. What actually makes Kubernetes difficult to approach

Basic Control Loop Workflow:

  1. Declare your desired state
  2. Kubernetes checks to see if current state is desired
  3. If not in desired state, controller(s) make or request changes to correct this

Kubernetes Cluster Infrastructure

Control Plane

  • Runs infrastructure controlling components
  • K8s API Server
    • front-end for control plane
    • central point of communication for all cluster objects
  • Controller Manager & Cloud Controller Manager
    • manage all controllers
  • Scheduler
    • assigns workloads to the underlying nodes
  • ETCD
    • stores all of K8s backing cluster data (state of objects, name of objects, …)

Worker Nodes

  • Kubelet
    • something like a K8s agent that runs on each node
    • uses container runtime interface
  • Kube-proxy
    • helps maintaining the networking rules on the underlying nodes
  • Any container runtime

Kubernetes Objects

Kubernetes Object YAMLs

  • apiVersion
  • kind
  • metadata
    • name
    • namespace
    • labels
    • annotations
  • spec

The Pod

kubernetes 1

  • Pods are the smallest deployable unit of computing that you can create and manage in Kubernetes
  • Pod is a Kubernetes construct
  • A pod can run multiple containers

Storage

  • Volumes
    • Ephemeral vs Persistent
  • Persistent Volumes
    • PersistentVolumeClaims
  • Container Storage Interface

Networking

  • Kubernetes Networking Services
    • ClusterIP
    • NodePort
    • LoadBalancer
    • ExternalName
  • Ingress (Contollers)

Workloads

  • DaemonSet
    • ensures a copy of a pod runs on every (or selected) node in the cluster
  • StatefulSet
    • manages stateful pods with stable identities and persistent storage
  • Deployment
    • manages stateless replicas of pods with easy scaling and updates

Namespaces

  • A virtual cluster within a single physical cluster that isolates resources like pods, services, and deployments, allowing multiple teams or projects to share the same cluster without interfering with each other
  • help with:
    • resource isolation
    • access control
    • organizing resources

Extending Kubernetes

  • Custom Resource Definitions
  • Operator Framework

Hamburger

kubernetes 2

Elastic’s Operator

  • kubectl get elasticsearch
  • YOUR controllers, built on top of THEIR controllers, making the entire stack happen

ECK

  • ECK (Elastic Cloud on Kubernetes) is an operator that lets Kubernetes manage Elasticsearch, Kibana, and other Elastic Stack components
  • It extends Kubernetes with custom resources so these services can be deployed, scaled, and upgraded like native workloads
  • This makes running and managing Elastic Stack on Kubernetes simple and declarative

Further Reads