Exploring the importance of Continuous Integration (CI) for successful software projects

Image from Devonblog.com

Hello readers, in this article, we are going to explore the importance of Continuous Integration for successful software projects, along with the benefits it can provide.

I will start by describing the concept of DevOps and then the purpose and importance of Continuous Integration and software development, and its benefits. I will then explore best practices for using continuous integration and in the end, I will put all the pieces together.

Before I start I would like to thanks Mr. Paul M. Duvall, Steve Matyas, and Andrew Glover for their amazing book that helped me a lot with building this article.

The Concept of DevOps

DevOps stand for development and operations. Now traditionally these have been separate roles, but in today’s computing world of rapid software delivery, DevOps has emerged as a role that bridges the gap between development and operations. DevOps concepts bridge the gap between the developer’s code and the deployment of that code to production environments with three main concepts, Continuous Delivery, Continuous Integration, and Continuous Deployment. These three concepts are often shortened to the acronym CI/CD. These concepts use automation to deliver the developer’s code to the production environment continuously and with high quality. Let us just clear up any confusion about Continuous Delivery versus Continuous Deployment before we start with the concept of CI. so let’s take a quick look at both of them.

Continuous Delivery: means that each and every code change goes through a process to make it ready to deploy to production, in other words, it is always delivery-ready. To get there, yeah change goes through the following steps.

  • It’s vetted via code reviews.
  • Issues are debugged by the developers.
  • An exhaustive series of automated tests are completed, such as unit tests and integration tests.
  • The code change is usually deployed to non-production environments, possibly for further validation.

Continuous Deployment: Continuous Deployment is differentiated from Continuous Delivery by one very important detail, with continuous deployment, deployment of every code change to production is automated and does not wait for management decision, this means that once a code change has gone through the process of vetting, automated testing, and debugging, it is considered truly ready for production and automatically deployed.

Now both continuous delivery and continuous deployment have as a sub-process continuous integration.

The Concept of Continuous integration (CI)

Continuous integration, or just CI is a software development practice under which developers regularly merge incremental code changes into a single shared code repository, this occurs daily or even several times per day.
The repository has a main branch that is considered to be always delivery-ready, now multiple developers working on the same repository can get messy unless it is managed properly and that's where continuous integration comes in. Developers typically work on feature branches off the main branch, and when the feature is ready, they request a code review, when reviewers approve the code, the code is merged with the main branch, and an automated build of the main branch is fired, this automated build builds the code, thus verifying that the code is buildable, it also runs test suites, such as unit tests for further validation, if anything goes wrong, the developers are immediately notified that the build is broken and they must fix it immediately, This rapid feedback ensures that the main branch remains clean, and allows multiple developers to work on a single code base efficiently.

How to set up Continuous Integration

First, we start with a code version control repository, in other words, you must use a version control repository in order to perform CI. In fact, even if you do not use CI, a version control repository should be standard for your project. The purpose of a version control repository is to manage changes to source code and other software assets using a controlled access repository[1]. Examples of the version control repository, you could use Github, Gerrit, Bitbucket, or one of the many others that are available.
Next, we choose a continuous integration delivery service that will automatically run our builds and the automated tests whenever a commit occurs within the repository and deploys the builds where required, you also need to create and configure your automated tests that will run when your builds get triggered as well.

Benefits of Continous Integration

Let us talk about the benefits of CI and how it has changed the way that we develop software. One of the benefits that it gives us is a shortened life cycle and product deployment, improved code quality and test coverage, stable production-ready code, and a consistent and repeatable build process. Let us dive into these a little bit deeper and explore these benefits.

  1. Quicker life cycle and product deployment

So with CI, we have a quicker development life cycle and this is because everything is automated, planned, because of the shortened life cycle and shortened development, we also have faster product deployment. So we have got incremental code changes that are constantly being tested and deployed.

2. Stable, production-ready code

We have got stable, production-ready code. So we have incremental changes that have been merged into the code repository, and the automated build is triggered, automated tests run. So we are at the point where we have that one-click deployment, or we can automatically deployed to production. But the fact is that we have got stable code that point that is ready. Since everything is automated, and since we are focused on incremental changes and testing them from the very start, we see a decrease in integration and testing time.

3. Code Quality and Testing Coverage

Adopting the continuous integration practice can also lead to increased code testing coverage, in fact, continuous integration servers can report test coverage. So we have an idea from the very start what kind of test coverage we have got, so we can use that information and improve, and that is part of what continuous integration is all about as well, because of all the automated testing, we see a decrease in manual testing, because we have automated unit integration tests. We have got those tests automated right in our build. Therefore, more bugs are detected way before production deployment. So at the point that we integrate the code, we fire off that build process. Well, we have automated unit integration tests running, and most of the bugs will be found hopefully at this point.[2]

Continuous Integration Best Practices

So let us discuss CI best practices include using version control tools, performing daily mainline commits, using virtual test environments, automating and releasing frequently, and displaying build reports and logs. Let us look at these in a little bit more detail.

  1. Use Version Control Tools

There are a number of options here, including open source tools. And using version control tools allows among other things, developers to develop along multiple branches, and the version control tools themselves do a tremendous job of managing and storing all the solution files for an entire software application, which is a monumental job. And there are multiple options, again, to run version control tools either on-prem locally or in the cloud.

2. Daily Mainline Commits

We should perform daily mainline commits. So integrated small changes. Pass the build test and commit the changes daily or even more frequently than daily. The smaller the changes, the quicker the builds, the more frequently we can deploy the production.

3. Virtual Test Environments

We can clone very sophisticated and complex production environments at a fraction of the cost. And we always get to start with a fresh environment. One of the nice things about virtual test environments is that you can revert back to a previous snapshot of your environment or a base state. That way you can avoid data or changes from previous tests leaking into a current test iteration. Another benefit of virtual test environments is creating environments on-demand, something you may hear is infrastructure as code. What this refers to is codifying your infrastructure. So for example in tools like Jenkins, Puppet or Chef or Terraform, you can create scripts, which will not only spin up the test environment but can configure the environment with specific configurations, not only that, if you codify your environment, that means that you can also test it just like any other code.

4. Automate and Release Often

Another best practice is automating and releasing often. When we release, we can release to multiple environments and we use release scripts for consistency, everything is programmed with continuous integration and everything is automated, this also includes the release scripts, which makes them consistent, easy to change, easy to test. And one of the environments that we want to release to is a test environment or a staging environment, and this is particularly important when there are additional tests that have to be run, for example, usability tests, UI tests. So if these tests need to be performed before we deploy to production then we need to release to tests. And an excellent practice involves automating rollbacks. As you know, nothing is perfect and bugs will slip through. So with help from automated rollback, we can ever to the last known good release[1].

5. Build reports and Logs

We want to track everything in logs because it is through logs that we can actually figure out what happened and how we can address it. We want to use reports to display and track log data and ensure that everyone has access to those reports. This is in the interest of visibility, transparency, and collaboration.

Putting It All Together

So, by adopting Continuous integration, There is a reduced chance of breaking the code base, because if you have a stable code base, there are fewer code conflicts because merging code is considerably easier, besides that, the code is validated and tested during integration. There is improved code quality, as you know it is impossible to keep track of everything in any modern software application, there are a lot of moving parts, interdependencies, and so on. By using a reliable process, one that systematically addresses all the requirements, knows about the dependencies, all the testing steps, and so on, we have a process that naturally supports quality. Another factor to improve code quality is the fact that automated testing is performed with each commit with the support automation tools, and integrating code daily or even more frequently finds bugs faster with faster feedback, and the cascading effect is better quality across the board.

Some common continuous integration mistakes include having a slow deployment pipeline, and to avoid that, we need to simplify the pipeline process. Pipelines with too many steps tend to be overcomplicated and can end up resulting in long waits between code commits and deployment to production. so-called short, wide deployment pipelines offer flexibility, enabling a decision to be made regarding which tests to run based on the sort of code change that was committed. This ensures that rapid feedback occurs, offering teams the ability to act efficiently to resolve failed tests.

That’s it, well done for reaching the end of this article.

References

[1] Continuous Integration: Improving Software Quality and Reducing Risk. by Paul M. Duvall , Steve Matyas, and Andrew Glover

[2] Learning Continuous Integration with Jenkins: A beginner’s guide to implementing Continuous Integration and Continuous Delivery using Jenkins 2, 2nd Edition by Nikhil Pathania

— Maher;

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store