Kubernetes Patterns - Examples

This GitHub project contains the examples from Kubernetes Patterns - Reusable Elements for Designing Cloud-Native Applications book by Bilgin Ibryam and Roland Huß

Installation instructions for the example prerequisites are summarised in INSTALL. By default, you need access to a vanilla Kubernetes installation, like Minikube. However, if add-ons are required, the example description explains this additional requirement.

For feedback, issues, or questions in general, please use the issue tracker to open cases. Also, we love contributions like spelling fixes, bug fixes, and improvements;…​ Please open pull requests; we are happy to review them!

Patterns

All examples are categorised according to the Book’s patterns category. Each example is contained in a different directory per pattern and is self-contained. ^[1]

Foundational Patterns

Predictable Demands

Our sample random generator deals with hard requirements on ConfigMap and PersistentVolumeClaims and resource limits.

Declarative Deployment

Rolling and fixed update of the random generator Deployment from version 1.0 to 2.0.

Health Probe

Liveness and Readiness probes for the random generator.

Managed Lifecycle

postStart and preStop hooks are demonstrated with the random generator application.

Automated Placement

These examples demonstrate a node selector, pod and node affinity, taint, and tolerations.

Behavioral Patterns

Batch Job

Generate thousands of random numbers into a file with a batch job

Periodic Job

Reuses the link: behavioral/BatchJob/README.adoc[Batch Job] example, but runs it periodically at a configured schedule

Daemon Service

Sample maintenance script for maintenance jobs on every node of a cluster

Singleton Service

Example of a PodDisruptionBudget for controlling voluntary disruptions

Stateless Service

Example for a stateless service deployed with a ReplicaSet

Stateful Service

StatefulSet example for our random-generator

Service Discovery

Various ways how to access our random-generator REST service

Self Awareness

Using the Downward API for setting environment variables and mounting resource fields as files with the random-generator service.

Structural Patterns

Init Container

Initialize an HTTP server’s HTML source from a remote git repository

Sidecar

Git polling example for a sidecar

Adapter

Adapter for exporting timing information from the sample random-generator application in a Prometheus format

Ambassador

Ambassador for moving on the log of the random-generator

Configuration Patterns

EnvVar Configuration

A simple example of how to use environment variables literally and from ConfigMaps and Secrets for our random-generator service.

Configuration Resource

Example of how to configure the random-generator Spring Boot application with a ConfigMap

Immutable Configuration

Several examples show how to use immutable configuration containers for application configuration. This directory includes examples of the simple Docker case and Kubernetes.

Configuration Template

This example shows how to use a template configuration standalone.xml, which is processed with a template processed and filled with data from ConfigMap before a Wildfly server is started.

Security Patterns

Process Containment

Example of how you can restrict the privileges of containers when running in Kubernetes.

Network Segmentation

This example shows you how you can restrict network access to Pods via Network Policies and how to control HTTP traffic with Istio’s Authentication Policies.

Secure Configuration

Examples show various techniques for storing configuration data securely in a Kubernetes Cluster.

Access Control

Example for Kubernetes' RBAC access control subsystem for controlling the access to the Kubernetes API server.

Advanced Patterns

Controller

A simple, pure shell-based controller watches ConfigMap resources for changes and restarts Pods using a label selector provided as an annotation. An additional example controller exposes an Ingress route when it detects an exposeService label attached to the service.

Operator

Operator based on the ConfigMap watch controller and introduces a CRD ConfigWatcher, which connects a ConfigMap with a set of Pods to restart in case of a config change.

Elastic Scale

Horizontal and vertical scaling examples with the random-generator service [*]

Image Builder

These examples demonstrate the setup of a chained build on OpenShift and use Knative build for doing builds within the cluster.

This work is licensed under a Creative Commons Attribution 4.0 International License

---

1. Examples marked with an "*" are functional but still lack the instructions for running the examples. If you feel adventurous, please try out the provided resource files yourself. Also happy to take PRs. Did we already mention that we love contributions? ;)