query-orchestration/cmd/metricsExample_test/main.go

// This main.go serves as an example of how to use the custom Prometheus metrics package.
//
// It demonstrates:
//
// 1. Setting up metrics with proper context handling
//
// 2. Configuring different types of metrics (Counter, Gauge, Histogram, Summary)
//
// 3. Recording metrics safely in a concurrent environment
//
// 4. Implementing graceful shutdown
//
// 5. Proper error handling
package main

import (
	"context"
	"fmt"
	"log"
	"math/rand"
	"os"
	"os/signal"
	"syscall"
	"time"

	"queryorchestration/internal/serviceconfig/build"

	"queryorchestration/internal/serviceconfig/observability/prometheus"
)

func getMetricsConfig() prometheus.MetricsConfig {
	standardDefinitions := prometheus.NewStandardMetricDefinitions()
	// Define metrics configuration using StandardMetricDefinitions
	config := prometheus.MetricsConfig{
		Port:      8080,
		Namespace: "myapp",
		Subsystem: "api",
		ServeHttp: true,
		// Start with standard metrics and append our custom one
		Definitions: append(standardDefinitions,
			prometheus.MetricDefinition{
				Type: prometheus.Counter,
				Name: "test_custom_events_total",
				Help: "Total number of custom test events",
				Labels: []string{
					"event_type",
					"priority",
				},
			},
		),
	}

	return config
}

func main() {
	rawVersion := build.GetVersion()
	fmt.Printf("Version: %s\n", rawVersion)
	timeVersion := build.GetVersionTimestamp()
	fmt.Printf("Version Timestamp: %s\n", timeVersion.String())

	ctx, cancel := context.WithCancel(context.Background())
	defer cancel()

	// Define metrics configuration using StandardMetricDefinitions
	config := getMetricsConfig()

	metrics, err := prometheus.New(ctx, config)
	if err != nil {
		log.Fatalf("Failed to create metrics: %v", err)
	}

	go simulateMetrics(ctx, metrics)

	sigChan := make(chan os.Signal, 1)
	signal.Notify(sigChan, syscall.SIGINT, syscall.SIGTERM)

	fmt.Println("Running... Press Ctrl+C to exit")
	<-sigChan
	fmt.Println("\nShutting down...")

	cancel()

	shutdownCtx, shutdownCancel := context.WithTimeout(context.Background(), 5*time.Second)
	defer shutdownCancel()

	if err := metrics.Shutdown(shutdownCtx); err != nil {
		log.Printf("Error shutting down metrics server: %v", err)
	}
}

// simulateMetrics demonstrates how to record different types of metrics
func simulateMetrics(ctx context.Context, metrics *prometheus.Metrics) {
	// Sample data for simulation
	endpoints := []string{"/api/users", "/api/products", "/api/orders"}
	methods := []string{"GET", "POST", "PUT", "DELETE"}
	statuses := []string{"200", "400", "500"}
	errorTypes := []string{"none", "validation", "database", "timeout"}
	dependencies := []string{"database", "cache", "auth-service", "email-service"}
	cacheOperations := []string{"hit", "miss", "set", "eviction"}
	taskTypes := []string{"cleanup", "indexing", "backup", "sync"}
	taskStatuses := []string{"running", "queued", "failed"}
	results := []string{"success", "failure", "timeout"}
	components := []string{"api", "database", "cache", "auth"}
	queueNames := []string{"main-queue", "background-queue", "notification-queue"}
	resourceTypes := []string{"cpu", "memory", "disk", "network"}

	// Custom metric sample data
	eventTypes := []string{"scheduled", "triggered", "manual"}
	priorities := []string{"low", "medium", "high"}

	ticker := time.NewTicker(2 * time.Second)
	defer ticker.Stop()

	for {
		select {
		case <-ctx.Done():
			return
		case <-ticker.C:
			// Simulate standard metrics

			// 1. Requests total
			endpoint := endpoints[rand.Intn(len(endpoints))]
			method := methods[rand.Intn(len(methods))]
			status := statuses[rand.Intn(len(statuses))]
			errorType := errorTypes[rand.Intn(len(errorTypes))]

			err := metrics.RecordRequestsTotal(1.0, endpoint, method, endpoint, status, errorType)
			if err != nil {
				log.Printf("Error recording requests metric: %v", err)
			}

			// 2. Request duration
			duration := rand.Float64() * 3 // Random duration between 0 and 3 seconds
			err = metrics.RecordRequestDurationSeconds(duration, endpoint, method, endpoint, status)
			if err != nil {
				log.Printf("Error recording duration metric: %v", err)
			}

			// 3. Active operations
			activeOps := rand.Float64() * 50
			err = metrics.RecordActiveOperations(activeOps, method)
			if err != nil {
				log.Printf("Error recording active operations metric: %v", err)
			}

			// 4. Resource usage
			resourceType := resourceTypes[rand.Intn(len(resourceTypes))]
			err = metrics.RecordResourceUsage(rand.Float64()*100, resourceType, "main-instance")
			if err != nil {
				log.Printf("Error recording resource usage metric: %v", err)
			}

			// 5. Dependency health
			dep := dependencies[rand.Intn(len(dependencies))]
			health := rand.Float64() // Random health score between 0 and 1
			err = metrics.RecordDependencyHealth(health, dep, "external_service")
			if err != nil {
				log.Printf("Error recording dependency health metric: %v", err)
			}

			// 6. Cache operations
			cacheOp := cacheOperations[rand.Intn(len(cacheOperations))]
			result := results[rand.Intn(len(results))]
			err = metrics.RecordCacheOperationsTotal(1.0, cacheOp, "main-cache", result)
			if err != nil {
				log.Printf("Error recording cache operations metric: %v", err)
			}

			// 7. Cache operation duration
			err = metrics.RecordCacheOperationDurationSeconds(rand.Float64()*0.1, cacheOp, result)
			if err != nil {
				log.Printf("Error recording cache duration metric: %v", err)
			}

			// 8. Background tasks
			taskType := taskTypes[rand.Intn(len(taskTypes))]
			taskStatus := taskStatuses[rand.Intn(len(taskStatuses))]
			err = metrics.RecordBackgroundTasks(rand.Float64()*10, taskType, taskStatus)
			if err != nil {
				log.Printf("Error recording background tasks metric: %v", err)
			}

			// 9. Active connections
			err = metrics.RecordActiveConnections(rand.Float64()*100, "http")
			if err != nil {
				log.Printf("Error recording active connections metric: %v", err)
			}

			// 10. Queue length
			queueName := queueNames[rand.Intn(len(queueNames))]
			err = metrics.RecordQueueLength(rand.Float64()*50, queueName)
			if err != nil {
				log.Printf("Error recording queue length metric: %v", err)
			}

			// 11. Errors total
			component := components[rand.Intn(len(components))]
			err = metrics.RecordErrorsTotal(1.0, component, errorType)
			if err != nil {
				log.Printf("Error recording errors total metric: %v", err)
			}

			// 12. Items in progress
			err = metrics.RecordItemsInProgress(rand.Float64()*20, endpoint, component)
			if err != nil {
				log.Printf("Error recording items in progress metric: %v", err)
			}

			// Custom metric (still using RecordMetric since it's not part of the generated stubs)
			eventType := eventTypes[rand.Intn(len(eventTypes))]
			priority := priorities[rand.Intn(len(priorities))]
			err = metrics.RecordMetric("test_custom_events_total", 1, map[string]string{
				"event_type": eventType,
				"priority":   priority,
			})
			if err != nil {
				log.Printf("Error recording custom event metric: %v", err)
			}
		}
	}
}