Integration Testing with .NET Aspire + xUnit
When to Use This Skill
Use this skill when:
- Writing integration tests for .NET Aspire applications
- Testing ASP.NET Core apps with real database connections
- Verifying service-to-service communication in distributed applications
- Testing with actual infrastructure (SQL Server, Redis, message queues) in containers
- Combining Playwright UI tests with Aspire-orchestrated services
- Testing microservices with proper service discovery and networking
Reference Files
- advanced-patterns.md [blocked]: Endpoint discovery, database testing, Playwright, conditional config, Respawn, service communication, message queues
- ci-and-tooling.md [blocked]: CI/CD integration, custom resource waiters, Aspire CLI with MCP
Core Principles
- Real Dependencies - Use actual infrastructure (databases, caches) via Aspire, not mocks
- Dynamic Port Binding - Let Aspire assign ports dynamically (
127.0.0.1:0) to avoid conflicts - Fixture Lifecycle - Use
IAsyncLifetimefor proper test fixture setup and teardown - Endpoint Discovery - Never hard-code URLs; discover endpoints from Aspire at runtime
- Parallel Isolation - Use xUnit collections to control test parallelization
- Health Checks - Always wait for services to be healthy before running tests
High-Level Testing Architecture
Required NuGet Packages
CRITICAL: File Watcher Fix for Integration Tests
When running many integration tests that each start an IHost, the default .NET host builder enables file watchers for configuration reload. This exhausts file descriptor limits on Linux.
Add this to your test project before any tests run:
Pattern 1: Basic Aspire Test Fixture (Modern API)
Pattern 2: Using the Fixture in Tests
See advanced-patterns.md [blocked] for Endpoint Discovery, Database Testing, Playwright UI Tests, Conditional Resource Configuration, Respawn database reset, Service-to-Service Communication, and Message Queue testing patterns.
Common Patterns Summary
Tricky / Non-Obvious Tips
Best Practices
- Use
IAsyncLifetime- Ensures proper async initialization and cleanup - Share fixtures via collections - Reduces test execution time by reusing app instances
- Discover endpoints dynamically - Never hard-code localhost:5000 or similar
- Wait for health checks - Don't assume services are immediately ready
- Test with real dependencies - Aspire makes it easy to use real SQL, Redis, etc.
- Clean up resources - Always implement
DisposeAsyncproperly - Use meaningful test data - Seed databases with realistic test data
- Test failure scenarios - Verify error handling and resilience
- Keep tests isolated - Each test should be independent and order-agnostic
- Monitor test execution time - If tests are slow, consider parallelization
See ci-and-tooling.md [blocked] for GitHub Actions setup, custom resource waiters, and Aspire CLI/MCP integration.
Debugging Tips
- Run Aspire Dashboard - When tests fail, check the dashboard at
http://localhost:15888 - Use Aspire CLI with MCP - Let AI assistants query real application state
- Enable detailed logging - Set
ASPIRE_ALLOW_UNSECURED_TRANSPORT=truefor more verbose output - Check container logs - Use
docker logsto inspect container output - Use breakpoints in fixtures - Debug fixture initialization to catch startup issues
- Verify resource names - Ensure resource names match between AppHost and tests


