Automating Software Deployment
This book is 100% complete
Completed on 2018-10-26
About the Book
This book teaches a technically inclined reader how to automate the deployment process of self-written software using Open Source tools. Thus it introduces you to Continuous Delivery, a practice under the DevOps umbrella. It enables faster feedback cycles, which is an important aspect of agile software development.
It makes some assumptions on the technologies used:
- Debian GNU/Linux or derived distributions, such as Ubuntu, as the operating system
- aptly for repository management
- Ansible for deployment and configuration management
- Go Continuous Delivery for orchestrating the process
It covers the following topics:
- Why automate the release process?
- Requirements for automated deployments
- Deployments enviroments (test, staging, production, ...) and their function
- Architecture of a deployment system
- Packaging the software
- Distributing the software
- Deploying the software
- Orchestrating the whole process with Go Continuous Delivery
- Version handling
- Unit Testing
- Smoke Testing and Rolling Deployments
- Security Considerations
- State Management and managing schema changes.
It shows a step by step process for creating a deployment system, and features two example applications deployed this way.
About the Author
About the Contributors
Table of Contents
-
1. Introduction
- 1.1 Who this book is for
-
2. Why Automate Deployments?
- 2.1 Save Time
- 2.2 Shorter Release Cycles
- 2.3 Shorter Feedback Cycles
- 2.4 Reliability of Releases
- 2.5 More Architectural Freedom
- 2.6 Advanced Quality Assurance Techniques
- 2.7 Automating Deployments or Continuous Delivery?
- 2.8 Agile, DevOps?
-
3. Requirements for Automating Deployments
- 3.1 Version Control
- 3.2 Automated Tests
- 3.3 Cooperation Between Operators and Developers
- 3.4 Partial Control Over the Destination Environment
- 3.5 Summary
-
4. Automating Deployments: Simplistic Deployment with Git and Bash
- 4.1 Reality Kicks In
- 4.2 More Headaches
- 4.3 A Less Daunting Perspective
-
5. Environments for your Software
- 5.1 The Development Environment
- 5.2 One or More Testing Environments
- 5.3 How Many Environments Should I Use?
- 5.4 Summary
-
6. The Anatomy of a Deployment System
- 6.1 The Pipeline Architecture
- 6.2 Anti-Pattern: Separate Builds per Environment
- 6.3 Everything Hinges on the Packaging Format
- 6.4 Summary
-
7. The Example Project: Package Info
- 7.1 Source
-
8. Building a Debian Package
- 8.1 Getting Started with Packaging
-
8.2 The
controlFile - 8.3 Directing the Build Process
- 8.4 Using Systemd to Start the Application Server
- 8.5 Triggering the Build
- 8.6 Summary
-
9. Distributing Debian Packages
- 9.1 Signatures
- 9.2 Preparing the Repository
- 9.3 Automating Repository Creation and Package Addition
- 9.4 Serving the Repositories
- 9.5 Summary
-
10. Package Deployment
- 10.1 Ansible: A Primer
- 10.2 Deploying with Ansible
- 10.3 Summary
-
11. A Virtual Playground for Automating Deployments
- 11.1 Requirements and Resource Usage
- 11.2 Introducing Vagrant
- 11.3 Configuring the Machines
- 11.4 Summary
-
12. Building in the Pipeline with Go Continuous Delivery
- 12.1 About Go Continuous Delivery
- 12.2 Installation
- 12.3 Building in the Pipeline
- 13. Summary
-
14. Distributing and Deploying Packages in the Pipeline
- 14.1 Uploading in the Pipeline
- 14.2 Deploying in the Pipeline
- 14.3 Results
- 14.4 Going All the Way to Production
- 14.5 Achievement Unlocked: Basic Continuous Delivery
-
15. Pipeline Improvements
- 15.1 Rollbacks and Installing Specific Versions
- 15.2 Configuration Templates
- 15.3 Avoiding the Rebuild Stampede
- 15.4 Summary
-
16. A Second Example Project: python-matheval
- 16.1 Meet python-matheval
- 16.2 Integration in GoCD
- 16.3 Summary
-
17. Automated Tests and Rolling Releases
- 17.1 Unit Tests
- 17.2 Smoke Testing
- 17.3 Adding Smoke Tests to the Pipeline, and Rolling Releases
- 17.4 Summary
-
18. Security
- 18.1 The Dangers of Centralization
- 18.2 Time to Market for Security Fixes
- 18.3 Audits and Software Bill of Materials
- 18.4 Conclusions
-
19. State Management
- 19.1 Synchronization Between Code and Database Versions
- 19.2 Decoupling Application and Database Versions
-
20. What’s Next?
- 20.1 Improved Quality Assurance
- 20.2 Metrics
- 20.3 Infrastructure Automation
- 20.4 Conclusions
- Acknowledgements
- Notes
The Leanpub 45-day 100% Happiness Guarantee
Within 45 days of purchase you can get a 100% refund on any Leanpub purchase, in two clicks.
See full terms...
