Book Image

Modern DevOps Practices

By : Gaurav Agarwal
Book Image

Modern DevOps Practices

By: Gaurav Agarwal

Overview of this book

Containers have entirely changed how developers and end-users see applications as a whole. With this book, you'll learn all about containers, their architecture and benefits, and how to implement them within your development lifecycle. You'll discover how you can transition from the traditional world of virtual machines and adopt modern ways of using DevOps to ship a package of software continuously. Starting with a quick refresher on the core concepts of containers, you'll move on to study the architectural concepts to implement modern ways of application development. You'll cover topics around Docker, Kubernetes, Ansible, Terraform, Packer, and other similar tools that will help you to build a base. As you advance, the book covers the core elements of cloud integration (AWS ECS, GKE, and other CaaS services), continuous integration, and continuous delivery (GitHub actions, Jenkins, and Spinnaker) to help you understand the essence of container management and delivery. The later sections of the book will take you through container pipeline security and GitOps (Flux CD and Terraform). By the end of this DevOps book, you'll have learned best practices for automating your development lifecycle and making the most of containers, infrastructure automation, and CaaS, and be ready to develop applications using modern tools and techniques.
Table of Contents (19 chapters)
1
Section 1: Container Fundamentals and Best Practices
7
Section 2: Delivering Containers
15
Section 3: Modern DevOps with GitOps

Introduction to config management

Let's look at the traditional way of hosting and managing applications. We first create a virtual machine from physical infrastructure and then log in manually to virtual machines. We can then choose to either run a set of scripts or do the setup manually. At least that's what we are doing till now, even in this book.

There are a number of problems with this approach. Let's look at some of them:

  • If we set up the server manually, the process is not repeatable. For example, if we need to build another server with a similar configuration, we will have to repeat the entire process.
  • Even if we use scripts, the scripts themselves are not idempotent. This means they cannot identify the delta configuration and apply them only if it is needed.
  • Typical production environments consist of a huge number of servers, and therefore setting everything up manually is a labor-intensive task and adds to the toil. Software engineers should...