Learn about build-related Tooling in Azure DevOps

Learn about build-related Tooling in Azure DevOps

Azure DevOps provides a comprehensive suite of tools and features tailored for build-related tasks in CI/CD workflows. Below is an overview of the key components, use cases, and best practices for using Azure DevOps for build automation.

1. Build Pipelines

Azure Pipelines is the core service for automating builds.

Features

1. Pipeline Types:

   • YAML-based pipelines for configuration-as-code.

   • Classic pipelines for GUI-based setup.

2. Multi-Platform Support: Build projects targeting Windows, Linux, or macOS environments.

3. Containerization: Support for Docker builds and containerized environments.

4. Custom Agents: Ability to use Microsoft-hosted agents or self-hosted build agents.

Use Cases

• Building applications written in different languages (e.g., .NET, Java, Python, Node.js).

• Creating cross-platform applications or using containerized builds.

2. Build Tools Integration

Azure DevOps supports integration with a wide variety of build tools.

Examples by Ecosystem

• .NET: MSBuild, NuGet, and Coverlet for building, packaging, and code coverage.

• Java: Maven, Gradle, Ant for dependency management and builds.

• Python: pip, tox, pytest for testing and building.

• Node.js: npm, Yarn, and webpack for managing dependencies and bundling.

• C++: CMake, MSBuild for cross-platform builds.

Best Practices

• Install dependencies in isolated environments (e.g., virtual environments for Python or Docker containers for Node.js).

• Cache build dependencies using Azure Pipelines caching tasks.

3. Dependency Management

Efficient dependency management is key to reducing build times and ensuring reproducibility.

Features

• Caching: Use built-in caching to store dependencies like npm packages, Python wheels, or Gradle caches.

  xxxxxxxxxx
  4
  1
  - task: Cache@2
  2
  inputs:
  3
     key: npm | "$(Agent.OS)_npm"
  4
     path: $(Pipeline.Workspace)/.npm
  

• Artifact Storage: Publish build artifacts to Azure DevOps or external repositories (e.g., NuGet Gallery, PyPI, npm registry).

4. Build Agents

Azure Pipelines offers two types of agents for running builds:

• Microsoft-Hosted Agents: Pre-configured with popular build tools (e.g., .NET SDK, Java JDK, Docker).

• Self-Hosted Agents: Custom environments for specific tools or configurations.

Best Practices

• Use Microsoft-hosted agents for standard builds to save on maintenance.

• Use self-hosted agents when specialized software or hardware is required.

5. Build Metrics and Reports

Azure DevOps offers detailed metrics and insights for builds:

• Pipeline Statistics: View build success/failure rates and duration trends.

• Test Results and Coverage: Integrate with tools like Coverlet, JaCoCo, or Istanbul to report code coverage and test metrics.

• Logs and Artifacts: Collect logs and artifacts from every build step for debugging.

Example

Use the PublishTestResults and PublishCodeCoverageResults tasks to view test and coverage results in the pipeline summary.

6. Build Optimization Techniques

• Parallel Builds: Run jobs in parallel to reduce build times.

• Incremental Builds: Use tools that only rebuild changed components (e.g., Gradle's incremental build feature).

• Dependency Caching: Cache dependencies like node_modules or pip installations to speed up builds.

• Containerized Builds: Use Docker images to create reproducible build environments.

7. Build Security

Security is a critical aspect of build pipelines:

• Secret Management: Use Azure Key Vault to securely store and access credentials and tokens.

• Secure Dependencies: Integrate tools like Snyk or WhiteSource Bolt to scan for vulnerabilities in dependencies.

• Code Scanning: Use static code analysis tools like SonarQube or Veracode during the build process.

8. Key Build-Related Tools in Azure DevOps

9. Use Cases

CI/CD for Microservices

• Use Docker for building individual services.

• Leverage Kubernetes for deploying and testing microservices.

Monorepo Builds

• Use build filters to trigger builds for specific directories or services.

• Cache dependencies for faster builds.

Cross-Platform Applications

Use matrix builds to test across multiple environments (e.g., Windows, macOS, Linux).

10. Example Build Pipeline Configuration

A sample YAML pipeline for a .NET project:

xxxxxxxxxx
24
1
trigger:
2
- main
3
pool:
4
  vmImage: 'windows-latest'
5
steps:
6
- task: UseDotNet@2
7
  inputs:
8
    packageType: 'sdk'
9
    version: '6.x'
10
- script: dotnet restore
11
  displayName: 'Restore NuGet packages'
12
- script: dotnet build --configuration Release
13
  displayName: 'Build Solution'
14
- script: dotnet test --configuration Release --collect:"Code Coverage"
15
  displayName: 'Run Tests with Coverage'
16
- task: PublishTestResults@2
17
  inputs:
18
    testResultsFormat: 'JUnit'
19
    testResultsFiles: '**/TestResults/.trx'
20
    mergeTestResults: true
21
- task: PublishCodeCoverageResults@1
22
  inputs:
23
    codeCoverageTool: 'cobertura'
24
    summaryFileLocation: '$(System.DefaultWorkingDirectory)/TestResults/coverage.cobertura.xml'