Featured image

Introduction

In today’s fast-paced software development landscape, ensuring seamless, reliable, and repeatable deployments across multiple environments is critical. Azure Pipelines, part of Azure DevOps, offers powerful capabilities to automate your application deployment process through multi-stage release pipelines. This comprehensive guide dives into creating multi-stage deployments using Azure Pipelines Classic release pipelines, enriched with practical examples and best practices.

Whether you’re deploying an ASP.NET Core app, a containerized microservice, or any other application, mastering multi-stage deployments helps you maintain control, enforce quality gates, and accelerate delivery.

What Are Multi-Stage Deployments?

Multi-stage deployments allow you to define a sequence of environments (stages) such as Development, Testing, Staging, and Production within a single pipeline. Each stage can have its own deployment tasks, conditions, and approval gates, enabling robust control over your release process.

Azure Pipelines supports multi-stage deployments via YAML pipelines and Classic release pipelines. This article focuses on the Classic release pipeline approach, providing a detailed walkthrough.

Prerequisites

Before we begin, ensure you have:

  • An Azure DevOps organization. You can create one for free if you don’t have it yet.
  • An Azure DevOps project where your pipeline and code reside.
  • A Classic release pipeline with at least one stage defined. If you’re new, you can create one via the Azure DevOps portal.

Step 1: Setting Up Continuous Deployment Triggers

Continuous deployment (CD) triggers automate the creation of a new release whenever a new build artifact is available. This eliminates manual intervention and accelerates delivery.

How to Enable CD Triggers

  1. Log into your Azure DevOps organization and open your project.
  2. Navigate to Azure Pipelines > Releases.
  3. Select the release pipeline you want to configure and click Edit.
  4. In the Artifacts section, click the Continuous deployment trigger icon.
  5. Toggle it to Enable. This configures the pipeline to automatically create a release on new artifacts.
  6. Under the first stage, click the Pre-deployment conditions icon.
  7. Verify that the deployment trigger is set to After release to auto-deploy upon release creation.

Best Practice

Keep your CD triggers enabled for environments like Development and Testing to speed up feedback loops. For Production, consider using manual approvals or gates to safeguard stability.

Step 2: Adding and Configuring Multiple Stages

Expanding your pipeline with multiple stages lets you mirror your deployment flow, e.g., Dev → QA → Staging → Production.

Adding a New Stage

  1. In the release pipeline editor, select + Add > New stage.
  2. Name the stage appropriately, such as “QA” or “Production”.
  3. Click the Pre-deployment conditions icon on the new stage.
  4. Set the trigger to After stage and select the preceding stage (e.g., after Dev).

Configuring Deployment Tasks

  1. From the Tasks drop-down, select the new stage.
  2. Define deployment tasks. For example, if deploying an ASP.NET Core app to Azure App Service, use the Deploy Azure App Service task.
# Sample task snippet for deploying to Azure App Service
- task: AzureRmWebAppDeployment@4
  inputs:
    azureSubscription: 'YourAzureServiceConnection'
    appType: 'webApp'
    WebAppName: 'your-app-service-name'
    packageForLinux: '$(System.DefaultWorkingDirectory)/drop/**/*.zip'

Real-World Scenario

In a realistic enterprise scenario, your pipeline might have:

  • Dev Stage: Automatically deploys after each successful build.
  • QA Stage: Deploys after Dev with automated tests.
  • Staging Stage: Deploys after QA with pre-deployment approval.
  • Production Stage: Deploys after Staging with strict manual approvals.

Step 3: Implementing Pre-Deployment Approvals

Approvals act as quality gates ensuring deployments meet criteria before progressing.

How to Add Approvals

  1. In the release pipeline editor, select the Pre-deployment conditions icon on the target stage.
  2. Enable Pre-deployment approvals toggle.
  3. Enter the approvers’ Azure DevOps user accounts.
  4. Uncheck “The user requesting a release or deployment should not approve it” to avoid self-approval.
  5. Save your pipeline.

Best Practices for Approvals

  • Assign approvers who have domain knowledge or responsibility for the environment.
  • Use approvals for sensitive environments like Production.
  • Combine approvals with automated gates (e.g., automated tests, monitoring checks) for stronger control.

Step 4: Creating and Managing Releases

Though CD triggers create releases automatically, manual release creation is sometimes necessary for controlled deployments.

Creating a Manual Release

  1. Navigate to Azure Pipelines > Releases.
  2. Select your release pipeline and click Create release.
  3. Provide a descriptive name or notes.
  4. Confirm artifact versions.
  5. Click Create.

Monitoring Release Progress

  • After creation, the release summary page shows each stage’s deployment status.
  • Approvers receive notifications to approve or reject deployments.

Handling Approvals

Approvers can add comments and approve or reject the deployment requests directly from Azure DevOps UI or email notifications.

Pro Tip

Configure notifications and dashboards to keep your team informed about release progress and approval requests.

Step 5: Monitoring and Troubleshooting Deployments

Robust monitoring ensures early detection and resolution of deployment issues.

Viewing Deployment Logs

  1. On the release summary, hover over a stage and select Logs.
  2. View logs for individual tasks in the deployment.
  3. Download logs for offline analysis or auditing.

Debug Mode

Enable debug mode by adding the variable System.Debug set to true in your pipeline variables to get verbose logs.

Practical Tips

  • Regularly review deployment logs to identify flaky tasks or configuration drifts.
  • Use log downloads to create knowledge bases or support tickets.
  • Integrate Azure Monitor or Application Insights for end-to-end visibility.

Advanced Tips and Best Practices

  • Use Variables and Variable Groups: Centralize configuration and secrets management.
  • Leverage Templates for YAML Pipelines: Though this guide focuses on Classic, YAML pipelines offer code-as-configuration advantages.
  • Automate Environment Provisioning: Use Infrastructure as Code (IaC) tools like ARM templates or Terraform.
  • Implement Gates: Combine approvals with automated gates like querying monitoring systems or running security scans.
  • Secure Service Connections: Use least privilege principles for Azure service connections.
  • Audit and Compliance: Regularly export and review deployment logs and approvals.

Conclusion

Azure Pipelines multi-stage deployments enable teams to implement controlled, automated, and repeatable release processes. By setting up continuous deployment triggers, defining multiple deployment stages, enforcing pre-deployment approvals, and monitoring deployments effectively, you can significantly improve your software delivery pipeline’s reliability and speed.

Embrace these practices to reduce risk and deliver value continuously. Start experimenting with your pipelines today and evolve your DevOps maturity!

Related Resources

Author

Joseph Perez

This article was crafted based on official Microsoft Azure DevOps documentation and enriched with practical insights for intermediate to advanced DevOps practitioners.