Spring Boot Development Patterns
Spring Boot architecture and API patterns for scalable, production-grade services.
When to Activate
- Building REST APIs with Spring MVC or WebFlux
- Structuring controller → service → repository layers
- Configuring Spring Data JPA, caching, or async processing
- Adding validation, exception handling, or pagination
- Setting up profiles for dev/staging/production environments
- Implementing event-driven patterns with Spring Events or Kafka
REST API Structure
Repository Pattern (Spring Data JPA)
Service Layer with Transactions
DTOs and Validation
Exception Handling
Caching
Requires @EnableCaching on a configuration class.
Async Processing
Requires @EnableAsync on a configuration class.
Logging (SLF4J)
Middleware / Filters
Pagination and Sorting
Error-Resilient External Calls
Rate Limiting (Filter + Bucket4j)
Security Note: The X-Forwarded-For header is untrusted by default because clients can spoof it.
Only use forwarded headers when:
- Your app is behind a trusted reverse proxy (nginx, AWS ALB, etc.)
- You have registered
ForwardedHeaderFilteras a bean - You have configured
server.forward-headers-strategy=NATIVEorFRAMEWORKin application properties - Your proxy is configured to overwrite (not append to) the
X-Forwarded-Forheader
When ForwardedHeaderFilter is properly configured, request.getRemoteAddr() will automatically
return the correct client IP from the forwarded headers. Without this configuration, use
request.getRemoteAddr() directly—it returns the immediate connection IP, which is the only
trustworthy value.
Background Jobs
Use Spring’s @Scheduled or integrate with queues (e.g., Kafka, SQS, RabbitMQ). Keep handlers idempotent and observable.
Observability
- Structured logging (JSON) via Logback encoder
- Metrics: Micrometer + Prometheus/OTel
- Tracing: Micrometer Tracing with OpenTelemetry or Brave backend
Production Defaults
- Prefer constructor injection, avoid field injection
- Enable
spring.mvc.problemdetails.enabled=truefor RFC 7807 errors (Spring Boot 3+) - Configure HikariCP pool sizes for workload, set timeouts
- Use
@Transactional(readOnly = true)for queries - Enforce null-safety via
@NonNullandOptionalwhere appropriate
Remember: Keep controllers thin, services focused, repositories simple, and errors handled centrally. Optimize for maintainability and testability.


