Packt+ | Advance your knowledge in tech

You're reading from The DevOps 2.2 Toolkit Self-Sufficient Docker Clusters

Product type Paperback

Published in Mar 2018

Publisher Packt

ISBN-13 9781788991278

Length 360 pages

Edition 1st Edition

Tools

Docker

Concepts

DevOps

Author (1):

Viktor Farcic

View More author details

Table of Contents (23) Chapters

Title Page

Dedication

Contributor

Packt Upsell

Preface

1. Introduction to Self-Adapting and Self-Healing Systems FREE CHAPTER

2. Choosing a Solution for Metrics Storage and Query

3. Deploying and Configuring Prometheus

4. Scraping Metrics

5. Defining Cluster-Wide Alerts

6. Alerting Humans

7. Alerting the System

The four quadrants of a dynamic and self-sufficient system

8. Self-Healing Applied to Services

9. Self-Adaptation Applied to Services

10. Painting the Big Picture – The Self-Sufficient System Thus Far

11. Instrumenting Services

12. Self-Adaptation Applied to Instrumented Services

13. Setting Up a Production Cluster

14. Self-Healing Applied to Infrastructure

15. Self-Adaptation Applied to Infrastructure

16. Blueprint of a Self-Sufficient System

1. Other Books You May Enjoy

Leave a review - let other readers know what you think

Index

Creating a Docker for AWS cluster

In the book, The DevOps 2.1 Toolkit: Docker Swarm (https://www.amazon.com/dp/1542468914), I argued that the best way to create a Swarm cluster in AWS is with a combination of Packer (https://www.packer.io/) and Terraform (https://www.terraform.io/). One of the alternatives was to use Docker CE for AWS (https://store.docker.com/editions/community/docker-ce-aws). At that time Docker for AWS was too immature. Today, the situation is different. Docker for AWS provides a robust Docker Swarm cluster with most, if not all the services we would expect from it.

We'll create a Docker for AWS cluster and, while in progress, discuss some of its aspects.

Before we start creating a cluster, we should choose a region. The only thing that truly matters is whether a region of your choice supports at least three Availability Zones (AZ). If there's only one availability zone, we'll risk downtime if it would become unavailable. With two availability zones, we'd lose Docker manager...