Why use Azure Pipelines for releases?
Azure Pipelines | TFS 2018 | TFS 2017 | TFS 2015
In Microsoft Team Foundation Server (TFS) 2018 and previous versions, build and release pipelines are called definitions, service connections are called service endpoints, stages are called environments, and jobs are called phases.
Release pipelines in Azure Pipelines and Team Foundation Server (TFS 2015.2 and later) are an essential element of DevOps CI/CD that help your team continuously deliver software to your customers at a faster pace and with lower risk. You can fully automate the testing and delivery of your software in multiple stages all the way to production, or set up semi-automated processes with approvals and on-demand deployments.
Watch this video - see Azure Pipelines releases in action.
Decide if it suits your scenarios - use the simple checklist.
See how it works - get a basic understanding of the process.
Get started now - follow the steps to deploy your apps.
Are release pipelines appropriate for you?
Consider using release pipelines if:
You develop applications and need to deploy them regularly to any platform, public or private cloud services, or App stores. Azure Pipelines has many out-of-the-box tasks to deploy a variety of applications. If you cannot find an out-of-the-box task to deploy your application using Azure Pipelines, consider this: if you can script the deployment of your application using Shell scripts or PowerShell scripts, utilities such as Ant or Maven, batch files or EXE utilities, then you can deploy it using Azure Pipelines. It also integrates with third party deployment systems such as Chef and Docker.
You use a continuous integration (CI) system and are looking for a fully-fledged continuous delivery management or release system. Whether you use Azure Pipelines, TFS, or Jenkins as your CI system, you can set up release pipelines to automatically deploy new builds to multiple stages. Even if we do not yet support integration with your favorite CI system or artifact repository, you can still write custom tasks to download and deploy artifacts from it.
You need to track the progress of releases. If you use several stages for your tests, release pipelines help you monitor whether a release has been deployed and tested on each of these stages. It also tracks whether an issue fixed by a developer, or a product backlog item completed by your team, has been deployed to a specific stage.
You need control of the deployments. Release pipelines let you specify which users can change the configuration of an stage, or approve the release to be deployed into a particular stage. If there is a problem with your deployment, Azure Pipelines helps you roll back to a previous deployment, and provide all the logs in one place to help you debug the problem.
You need audit history for all releases and their deployments. Release pipelines provides a history of all changes to the pipelines, configurations, and deployments. Azure Pipelines also provides a history of all the activity performed during each deployment. Each release is accompanied by a listing of new features and developer commits that went into that release.
How do release pipelines work?
Release pipelines store the data about your pipelines, stages, tasks, releases, and deployments in Azure Pipelines or TFS.
Azure Pipelines runs the following steps as part of every deployment:
Pre-deployment approval: When a new deployment request is triggered, Azure Pipelines checks whether a pre-deployment approval is required before deploying a release to a stage. If it is required, it sends out email notifications to the appropriate approvers.
Queue deployment job: Azure Pipelines schedules the deployment job on an available automation agent. An agent is a piece of software that is capable of running tasks in the deployment.
Agent selection: An automation agent picks up the job. The agents for release pipelines are exactly the same as those that run your builds in Azure Pipelines and TFS. A release pipeline can contain settings to select an appropriate agent at runtime.
Download artifacts: The agent downloads all the artifacts specified in that release (provided you have not opted to skip the download). The agent currently understands two types of artifacts: Azure Pipelines artifacts and Jenkins artifacts.
Run the deployment tasks: The agent then runs all the tasks in the deployment job to deploy the app to the target servers for a stage.
Generate progress logs: The agent creates detailed logs for each step while running the deployment, and pushes these logs back to Azure Pipelines or TFS.
Post-deployment approval: When deployment to a stage is complete, Azure Pipelines checks if there is a post-deployment approval required for that stage. If no approval is required, or upon completion of a required approval, it proceeds to trigger deployment to the next stage.
At present, release pipelines and build pipelines have separate designer interfaces (separate UIs). The main differences in the pipelines are the support in release pipelines for different types of triggers, and the support for approvals and gates. However, you can represent a pipeline using YAML, in which case the pipeline can combine both release and build capabilities.
Get started now!
Simply follow these steps:
- What is a draft release?
- When and why would I abandon a release?
- Download Team Foundation Server
- Install and configure Team Foundation Server
- Sign up for Azure Pipelines
Help and support
We'd love to hear your thoughts. Choose the type you'd like to provide:
Our feedback system is built on GitHub Issues. Read more on our blog.