Skip to main content
SGIVU implements comprehensive observability through Spring Boot Actuator, Zipkin distributed tracing, and structured logging.

​
Actuator Endpoints

Spring Boot Actuator provides production-ready monitoring and management endpoints.

​
Endpoint Exposure

​
management.endpoints.web.exposure.include
string
required
Comma-separated list of actuator endpoints to expose over HTTP. Different profiles expose different endpoints.

​
Default Profile (All Services)

management:
  endpoints:
    web:
      exposure:
        include: health, info
By default, services only expose health and info endpoints to minimize the attack surface.

​
Development Profile

management:
  endpoints:
    web:
      exposure:
        include: "*"
  endpoint:
    health:
      show-details: always
Development profiles expose all actuator endpoints for debugging. Never use "*" in production.

​
Production Profile

management:
  endpoints:
    web:
      exposure:
        include: health, info, prometheus
Production adds the prometheus endpoint for metrics scraping by monitoring systems.

​
Exposed Endpoints

Health

GET /actuator/healthService health status and readiness checks

Info

GET /actuator/infoApplication metadata and build information

Prometheus

GET /actuator/prometheusMetrics in Prometheus format (production only)

All Endpoints

GET /actuatorLists all available endpoints (dev only)

​
Health Check Configuration

​
Health Details Visibility

​
management.endpoint.health.show-details
string
required
Controls health check detail visibility. Values: never, when-authorized, always.

​
Default and Production

management:
  endpoint:
    health:
      show-details: never
Services hide health details by default to prevent information leakage. Health endpoint returns only UP or DOWN status.

​
Development

management:
  endpoint:
    health:
      show-details: always
Development profiles show full health details including database connectivity, disk space, and other indicators.

​
Example Health Responses

​
Production (show-details: never)

{
  "status": "UP"
}

​
Development (show-details: always)

{
  "status": "UP",
  "components": {
    "db": {
      "status": "UP",
      "details": {
        "database": "PostgreSQL",
        "validationQuery": "isValid()"
      }
    },
    "diskSpace": {
      "status": "UP",
      "details": {
        "total": 499963174912,
        "free": 262958080000,
        "threshold": 10485760
      }
    },
    "ping": {
      "status": "UP"
    }
  }
}

​
Distributed Tracing

SGIVU uses Zipkin for distributed tracing to track requests across microservices.

​
Tracing Configuration

​
management.tracing.sampling.probability
number
required
Percentage of requests to trace (0.0 to 1.0). Set to 0.1 (10%) by default to balance observability with performance.
​
management.zipkin.tracing.endpoint
string
required
Zipkin server endpoint for sending trace data.
management:
  tracing:
    sampling:
      probability: 0.1
  zipkin:
    tracing:
      endpoint: http://sgivu-zipkin:9411/api/v2/spans

​
Sampling Strategy

Tracing every request can impact performance and generate massive amounts of data. A 10% sampling rate:
  • Provides sufficient visibility into system behavior
  • Captures enough traces to detect patterns and issues
  • Minimizes performance overhead
  • Reduces storage requirements
Increase the sampling rate to 1.0 (100%) in development for complete visibility.

​
Service Examples

​
Gateway Service

management:
  tracing:
    sampling:
      probability: 0.1
  zipkin:
    tracing:
      endpoint: http://sgivu-zipkin:9411/api/v2/spans

​
Auth Service

management:
  tracing:
    sampling:
      probability: 0.1
  zipkin:
    tracing:
      endpoint: http://sgivu-zipkin:9411/api/v2/spans

​
Domain Services (User, Client, Vehicle, Purchase-Sale)

All domain services use identical tracing configuration: 
management:
  tracing:
    sampling:
      probability: 0.1
  zipkin:
    tracing:
      endpoint: http://sgivu-zipkin:9411/api/v2/spans

​
Logging Configuration

SGIVU uses SLF4J with Logback for structured logging.

​
Log Levels

​
logging.level.root
string
Root logger level. Set to INFO in all environments for balanced logging.
​
logging.level.{package}
string
Package-specific log levels for fine-grained control.

​
Standard Configuration

logging:
  level:
    root: INFO

​
Service-Specific Logging

​
Gateway Service

logging:
  level:
    root: INFO
    com.sgivu.gateway.security: INFO
    com.sgivu.gateway.controller: INFO
The Gateway explicitly sets log levels for security and controller packages to ensure authentication/authorization events are captured.

​
Vehicle Service

logging:
  level:
    root: INFO
    software:
      amazon:
        awssdk: info
The Vehicle Service sets AWS SDK logging to info to reduce verbosity from S3 operations.

​
Log Level Guide

ERROR

Critical failures requiring immediate attention

WARN

Potentially harmful situations that don’t stop execution

INFO

Important runtime events and milestones

DEBUG

Detailed diagnostic information (development only)

​
Complete Observability Examples

​
Gateway Service

management:
  endpoints:
    web:
      exposure:
        include: health, info
  endpoint:
    health:
      show-details: never
  tracing:
    sampling:
      probability: 0.1
  zipkin:
    tracing:
      endpoint: http://sgivu-zipkin:9411/api/v2/spans

logging:
  level:
    root: INFO
    com.sgivu.gateway.security: INFO
    com.sgivu.gateway.controller: INFO

​
Auth Service

management:
  endpoints:
    web:
      exposure:
        include: health, info
  endpoint:
    health:
      show-details: never
  tracing:
    sampling:
      probability: 0.1
  zipkin:
    tracing:
      endpoint: http://sgivu-zipkin:9411/api/v2/spans

logging:
  level:
    root: INFO

​
User Service

management:
  endpoints:
    web:
      exposure:
        include: health, info
  endpoint:
    health:
      show-details: never
  tracing:
    sampling:
      probability: 0.1
  zipkin:
    tracing:
      endpoint: http://sgivu-zipkin:9411/api/v2/spans

logging:
  level:
    root: INFO

​
Client Service

management:
  endpoints:
    web:
      exposure:
        include: health, info
  endpoint:
    health:
      show-details: never
  tracing:
    sampling:
      probability: 0.1
  zipkin:
    tracing:
      endpoint: http://sgivu-zipkin:9411/api/v2/spans

logging:
  level:
    root: INFO

​
Vehicle Service

management:
  endpoints:
    web:
      exposure:
        include: health, info
  endpoint:
    health:
      show-details: never
  tracing:
    sampling:
      probability: 0.1
  zipkin:
    tracing:
      endpoint: http://sgivu-zipkin:9411/api/v2/spans

logging:
  level:
    root: INFO
    software:
      amazon:
        awssdk: info

​
Purchase-Sale Service

management:
  endpoints:
    web:
      exposure:
        include: health, info
  endpoint:
    health:
      show-details: never
  tracing:
    sampling:
      probability: 0.1
  zipkin:
    tracing:
      endpoint: http://sgivu-zipkin:9411/api/v2/spans

logging:
  level:
    root: INFO

​
Observability Stack

1

Service Instrumentation

Spring Boot Actuator and Micrometer automatically instrument services with metrics, health checks, and trace context propagation.
2

Trace Collection

Sampled requests generate trace spans that are sent to the Zipkin server at http://sgivu-zipkin:9411.
3

Metrics Export

Production services expose Prometheus metrics at /actuator/prometheus for scraping.
4

Health Monitoring

Orchestrators and load balancers query /actuator/health to determine service availability.
5

Log Aggregation

Container logs (stdout/stderr) are collected by the container runtime and can be shipped to centralized logging systems.

​
Monitoring Best Practices

Configure your orchestrator (Kubernetes, Docker Swarm, ECS) to:
  1. Perform health checks on /actuator/health
  2. Set appropriate timeout and interval values
  3. Restart unhealthy containers automatically
  4. Alert on repeated health check failures
Example Kubernetes liveness probe:
livenessProbe:
  httpGet:
    path: /actuator/health
    port: 8080
  initialDelaySeconds: 30
  periodSeconds: 10
  timeoutSeconds: 5
  failureThreshold: 3
Access the Zipkin UI (typically at http://localhost:9411) to:
  1. Search for traces by service name or time range
  2. Visualize request flows across services
  3. Identify performance bottlenecks
  4. Debug distributed transaction failures
  5. Analyze service dependencies
Low traffic environments (< 100 req/min):
management:
  tracing:
    sampling:
      probability: 1.0  # Trace 100%
Medium traffic environments (100-1000 req/min):
management:
  tracing:
    sampling:
      probability: 0.1  # Trace 10%
High traffic environments (> 1000 req/min):
management:
  tracing:
    sampling:
      probability: 0.01  # Trace 1%
  1. Deploy Prometheus to scrape /actuator/prometheus endpoints
  2. Configure Prometheus as a data source in Grafana
  3. Import Spring Boot dashboards from Grafana.com
  4. Create custom dashboards for business metrics
  5. Set up alerting rules in Prometheus/Alertmanager
Use structured logging (JSON format) for better log analysis:
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

Logger log = LoggerFactory.getLogger(UserService.class);

// Good: Structured with context
log.info("User created: userId={}, email={}", user.getId(), user.getEmail());

// Better: With MDC for request tracing
MDC.put("userId", user.getId());
log.info("User login successful");
This enables powerful log filtering and analysis in centralized logging systems.

​
Troubleshooting

Causes:
  • Actuator dependency not included
  • Endpoints not exposed in configuration
  • Context path misconfiguration
Solutions:
  1. Verify spring-boot-starter-actuator is in dependencies
  2. Check management.endpoints.web.exposure.include configuration
  3. Try full path: /actuator/health (not just /health)
Causes:
  • Zipkin server unreachable
  • Sampling probability too low
  • Micrometer tracing not configured
Solutions:
  1. Verify Zipkin is running: curl http://sgivu-zipkin:9411/health
  2. Increase sampling to 1.0 for testing
  3. Check service logs for Zipkin connection errors
  4. Verify spring-cloud-starter-zipkin dependency
Risk: Information disclosure and potential security vulnerabilitiesSolution: Ensure production profile uses restricted endpoint exposure:
management:
  endpoints:
    web:
      exposure:
        include: health, info, prometheus
Never use include: "*" in production.
Cause: Sampling probability too high for traffic volumeSolution: Reduce sampling probability:
management:
  tracing:
    sampling:
      probability: 0.01  # 1% instead of 10%
Monitor CPU usage after adjustment.

Spring Configuration

Server and application settings

Environment Variables

Observability-related environment variables

Deployment

Zipkin and monitoring stack deployment

Service Architecture

How observability fits into the system

Build docs developers (and LLMs) love

Get started for freeTalk to us