Azure Pipelines

We were using the Azure Pipelines for CI/CD for our product deployment on Azure. We have integrated automated Pipeline associated with GitHub so whenever we merge any PR or commit code, that code becomes live in our frontend/backend in just few minutes. We have done integration via the Docker image.

It is good tool if you are doing continuous improvements in your code and you wish it goes live whenever you push code to GitHub. So integrating Azure Pipeline, it automatically does CI/CD in the background once you push code/merge code and it is live in few minutes. It also does some automated tests if you have wrote scripts

We're using Azure Pipelines for automating deploying our changes from the code in the repository to our production environment. We have a few pipelines for our separate environments. We're checking the quality of the changes that are going to be made and then if everything passes, we're doing a deployment.

It's good when you want to have an easy way to deploy your changes to the production environment. Furthermore, there are a lot of possibilities to add some tasks from different services, so it is easy to fit the pipeline to the case you have to have. When you need to have good error handling it is less appropriate.

Azure Pipelines are used for any deployment of our apps, backend services and test automation. This is the backbone of our deployment process allows us to deliver within our release cycle. Our current deployment cycle is monthly - but at times we may have smaller more controlled deployments within a release cycle. Azure Pipelines are fully integrated with our workflows as we also use Azure DevOps and implementation was more effective and efficient for us than other tools that might exist.

With a fully Microsoft Azure based workflow - Azure Pipelines makes absolute sense. Azure Pipelines are robust and work very well with SonarQube for test coverage and are shared with our developers. This prevents the developers for pushing code without unit tests across our backend and frontend platforms. We have reduced our instances of manual regression tests especially when there are multiple teams working across the same repositories.