To expose custom metrics in Spring Boot, you can use Micrometer, which is the metrics collection facade integrated into Spring Boot Actuator. Spring Boot Actuator provides production-ready endpoints like /actuator/metrics, and you can extend it with custom metrics to track application-specific behavior.

Step-by-Step Guide to Expose Custom Metrics

1. Add Required Dependencies

If you’re using Spring Boot Starter Actuator (most common case):

<dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-actuator</artifactId>
</dependency>

If you want to export metrics to Prometheus, Datadog, etc., add the corresponding registry (e.g., Prometheus):

<dependency>
    <groupId>io.micrometer</groupId>
    <artifactId>micrometer-registry-prometheus</artifactId>
</dependency>

2. Enable Metrics Endpoint in application.properties

management.endpoints.web.exposure.include=metrics,prometheus

This exposes:

• /actuator/metrics – for all metrics
• /actuator/metrics/<metric-name> – for individual metrics
• /actuator/prometheus – for Prometheus scraping (if you added the Prometheus registry)

3. Inject MeterRegistry to Register Custom Metrics

You can use MeterRegistry to create custom counters, timers, gauges, etc.

Custom Counter Example:

@Component
public class CustomMetricsService {

    private final Counter userCreatedCounter;

    public CustomMetricsService(MeterRegistry registry) {
        this.userCreatedCounter = Counter.builder("app.user.created.count")
                .description("Number of users created")
                .register(registry);
    }

    public void userCreated() {
        // business logic to create a user
        userCreatedCounter.increment(); // increment the custom metric
    }
}

Now you can see this metric at:

/actuator/metrics/app.user.created.count

4. Expose Custom Timers (Duration Tracking)

@Component
public class CustomTimerService {

    private final Timer requestTimer;

    public CustomTimerService(MeterRegistry registry) {
        this.requestTimer = Timer.builder("app.request.duration")
                .description("Request processing time")
                .register(registry);
    }

    public void process() {
        requestTimer.record(() -> {
            // code whose duration you want to measure
            try {
                Thread.sleep(200); // simulate delay
            } catch (InterruptedException e) {
                Thread.currentThread().interrupt();
            }
        });
    }
}

Check metrics via:

/actuator/metrics/app.request.duration

5. Expose Gauges (Real-time Value Tracking)

For values that fluctuate, like queue size or memory usage.

@Component
public class QueueMonitor {

    private final Queue<String> queue = new ConcurrentLinkedQueue<>();

    public QueueMonitor(MeterRegistry registry) {
        Gauge.builder("app.queue.size", queue, Queue::size)
             .description("Size of in-memory queue")
             .register(registry);
    }

    public void addToQueue(String item) {
        queue.add(item);
    }

    public String removeFromQueue() {
        return queue.poll();
    }
}

Access via:

/actuator/metrics/app.queue.size

Types of Custom Metrics

Type	Description	Method
Counter	Increments when an event happens	Counter.builder(...).increment()
Gauge	Reports a changing value (e.g., queue size)	Gauge.builder(...).register()
Timer	Records duration of operations	Timer.record(Runnable)
Histogram	Shows distribution of values (via Timer)	Handled automatically in Timer