Merge pull request #568 from Beckn-One/feat/observability

Feat/observability

pkg/plugin/definition/metrics.go

@@ -0,0 +1,15 @@
+package definition
+
+import (
+	"context"
+
+	"github.com/beckn-one/beckn-onix/pkg/telemetry"
+)
+
+// OtelSetupMetricsProvider encapsulates initialization of OpenTelemetry metrics
+// providers. Implementations wire exporters and return a Provider that the core
+// application can manage.
+type OtelSetupMetricsProvider interface {
+	// New initializes a new telemetry provider instance with the given configuration.
+	New(ctx context.Context, config map[string]string) (*telemetry.Provider, func() error, error)
+}

pkg/plugin/implementation/cache/cache.go

@@ -7,7 +7,12 @@ import (
 	"os"
 	"time"
 
+	"go.opentelemetry.io/otel/attribute"
+	"go.opentelemetry.io/otel/metric"
+
 	"github.com/beckn-one/beckn-onix/pkg/log"
+	"github.com/beckn-one/beckn-onix/pkg/telemetry"
+	"github.com/redis/go-redis/extra/redisotel/v9"
 	"github.com/redis/go-redis/v9"
 )
 
@@ -31,7 +36,8 @@ type Config struct {
 
 // Cache wraps a Redis client to provide basic caching operations.
 type Cache struct {
-	Client RedisClient
+	Client  RedisClient
+	metrics *CacheMetrics
 }
 
 // Error variables to describe common failure modes.
@@ -77,26 +83,80 @@ func New(ctx context.Context, cfg *Config) (*Cache, func() error, error) {
 		return nil, nil, fmt.Errorf("%w: %v", ErrConnectionFail, err)
 	}
 
+	// Enable OpenTelemetry instrumentation for tracing and metrics
+	// This will automatically collect Redis operation metrics and expose them via /metrics endpoint
+	if redisClient, ok := client.(*redis.Client); ok {
+		if err := redisotel.InstrumentTracing(redisClient); err != nil {
+			// Log error but don't fail - instrumentation is optional
+			log.Debugf(ctx, "Failed to instrument Redis tracing: %v", err)
+		}
+
+		if err := redisotel.InstrumentMetrics(redisClient); err != nil {
+			// Log error but don't fail - instrumentation is optional
+			log.Debugf(ctx, "Failed to instrument Redis metrics: %v", err)
+		}
+	}
+
+	metrics, _ := GetCacheMetrics(ctx)
+
 	log.Infof(ctx, "Cache connection to Redis established successfully")
-	return &Cache{Client: client}, client.Close, nil
+	return &Cache{Client: client, metrics: metrics}, client.Close, nil
 }
 
 // Get retrieves the value for the specified key from Redis.
 func (c *Cache) Get(ctx context.Context, key string) (string, error) {
-	return c.Client.Get(ctx, key).Result()
+	result, err := c.Client.Get(ctx, key).Result()
+	if c.metrics != nil {
+		attrs := []attribute.KeyValue{
+			telemetry.AttrOperation.String("get"),
+		}
+		switch {
+		case err == redis.Nil:
+			c.metrics.CacheMissesTotal.Add(ctx, 1, metric.WithAttributes(attrs...))
+			c.metrics.CacheOperationsTotal.Add(ctx, 1,
+				metric.WithAttributes(append(attrs, telemetry.AttrStatus.String("miss"))...))
+		case err != nil:
+			c.metrics.CacheOperationsTotal.Add(ctx, 1,
+				metric.WithAttributes(append(attrs, telemetry.AttrStatus.String("error"))...))
+		default:
+			c.metrics.CacheHitsTotal.Add(ctx, 1, metric.WithAttributes(attrs...))
+			c.metrics.CacheOperationsTotal.Add(ctx, 1,
+				metric.WithAttributes(append(attrs, telemetry.AttrStatus.String("hit"))...))
+		}
+	}
+	return result, err
 }
 
 // Set stores the given key-value pair in Redis with the specified TTL (time to live).
 func (c *Cache) Set(ctx context.Context, key, value string, ttl time.Duration) error {
-	return c.Client.Set(ctx, key, value, ttl).Err()
+	err := c.Client.Set(ctx, key, value, ttl).Err()
+	c.recordOperation(ctx, "set", err)
+	return err
 }
 
 // Delete removes the specified key from Redis.
 func (c *Cache) Delete(ctx context.Context, key string) error {
-	return c.Client.Del(ctx, key).Err()
+	err := c.Client.Del(ctx, key).Err()
+	c.recordOperation(ctx, "delete", err)
+	return err
 }
 
 // Clear removes all keys in the currently selected Redis database.
 func (c *Cache) Clear(ctx context.Context) error {
 	return c.Client.FlushDB(ctx).Err()
 }
+
+func (c *Cache) recordOperation(ctx context.Context, op string, err error) {
+	if c.metrics == nil {
+		return
+	}
+	status := "success"
+	if err != nil {
+		status = "error"
+	}
+	c.metrics.CacheOperationsTotal.Add(ctx, 1,
+		metric.WithAttributes(
+			telemetry.AttrOperation.String(op),
+			telemetry.AttrStatus.String(status),
+		))
+}

pkg/plugin/implementation/cache/cache_metrics.go

@@ -0,0 +1,69 @@
+package cache
+
+import (
+	"context"
+	"fmt"
+	"sync"
+
+	"go.opentelemetry.io/otel"
+	"go.opentelemetry.io/otel/metric"
+)
+
+// CacheMetrics exposes cache-related metric instruments.
+type CacheMetrics struct {
+	CacheOperationsTotal metric.Int64Counter
+	CacheHitsTotal       metric.Int64Counter
+	CacheMissesTotal    metric.Int64Counter
+}
+
+var (
+	cacheMetricsInstance *CacheMetrics
+	cacheMetricsOnce     sync.Once
+	cacheMetricsErr      error
+)
+
+// GetCacheMetrics lazily initializes cache metric instruments and returns a cached reference.
+func GetCacheMetrics(ctx context.Context) (*CacheMetrics, error) {
+	cacheMetricsOnce.Do(func() {
+		cacheMetricsInstance, cacheMetricsErr = newCacheMetrics()
+	})
+	return cacheMetricsInstance, cacheMetricsErr
+}
+
+func newCacheMetrics() (*CacheMetrics, error) {
+	meter := otel.GetMeterProvider().Meter(
+		"github.com/beckn-one/beckn-onix/cache",
+		metric.WithInstrumentationVersion("1.0.0"),
+	)
+
+	m := &CacheMetrics{}
+	var err error
+
+	if m.CacheOperationsTotal, err = meter.Int64Counter(
+		"onix_cache_operations_total",
+		metric.WithDescription("Redis cache operations"),
+		metric.WithUnit("{operation}"),
+	); err != nil {
+		return nil, fmt.Errorf("onix_cache_operations_total: %w", err)
+	}
+
+	if m.CacheHitsTotal, err = meter.Int64Counter(
+		"onix_cache_hits_total",
+		metric.WithDescription("Redis cache hits"),
+		metric.WithUnit("{hit}"),
+	); err != nil {
+		return nil, fmt.Errorf("onix_cache_hits_total: %w", err)
+	}
+
+	if m.CacheMissesTotal, err = meter.Int64Counter(
+		"onix_cache_misses_total",
+		metric.WithDescription("Redis cache misses"),
+		metric.WithUnit("{miss}"),
+	); err != nil {
+		return nil, fmt.Errorf("onix_cache_misses_total: %w", err)
+	}
+
+	return m, nil
+}
+
+

pkg/plugin/implementation/otelsetup/cmd/plugin.go

@@ -0,0 +1,79 @@
+package main
+
+import (
+	"context"
+	"errors"
+	"strconv"
+	"time"
+
+	"github.com/beckn-one/beckn-onix/pkg/log"
+	"github.com/beckn-one/beckn-onix/pkg/plugin/implementation/otelsetup"
+	"github.com/beckn-one/beckn-onix/pkg/telemetry"
+)
+
+// metricsProvider implements the OtelSetupMetricsProvider interface for the otelsetup plugin.
+type metricsProvider struct {
+	impl otelsetup.Setup
+}
+
+// New creates a new telemetry provider instance.
+func (m metricsProvider) New(ctx context.Context, config map[string]string) (*telemetry.Provider, func() error, error) {
+	if ctx == nil {
+		return nil, nil, errors.New("context cannot be nil")
+	}
+
+	// Convert map[string]string to otelsetup.Config
+	telemetryConfig := &otelsetup.Config{
+		ServiceName:    config["serviceName"],
+		ServiceVersion: config["serviceVersion"],
+		Environment:    config["environment"],
+		MetricsPort:    config["metricsPort"],
+	}
+
+	// Parse enableMetrics as boolean
+	if enableMetricsStr, ok := config["enableMetrics"]; ok && enableMetricsStr != "" {
+		enableMetrics, err := strconv.ParseBool(enableMetricsStr)
+		if err != nil {
+			log.Warnf(ctx, "Invalid enableMetrics value '%s', defaulting to true: %v", enableMetricsStr, err)
+			telemetryConfig.EnableMetrics = true
+		} else {
+			telemetryConfig.EnableMetrics = enableMetrics
+		}
+	} else {
+		telemetryConfig.EnableMetrics = true // Default to true if not specified or empty
+	}
+
+	// Apply defaults if fields are empty
+	if telemetryConfig.ServiceName == "" {
+		telemetryConfig.ServiceName = otelsetup.DefaultConfig().ServiceName
+	}
+	if telemetryConfig.ServiceVersion == "" {
+		telemetryConfig.ServiceVersion = otelsetup.DefaultConfig().ServiceVersion
+	}
+	if telemetryConfig.Environment == "" {
+		telemetryConfig.Environment = otelsetup.DefaultConfig().Environment
+	}
+
+	log.Debugf(ctx, "Telemetry config mapped: %+v", telemetryConfig)
+	provider, err := m.impl.New(ctx, telemetryConfig)
+	if err != nil {
+		log.Errorf(ctx, err, "Failed to create telemetry provider instance")
+		return nil, nil, err
+	}
+
+	// Wrap the Shutdown function to match the closer signature
+	var closer func() error
+	if provider != nil && provider.Shutdown != nil {
+		closer = func() error {
+			shutdownCtx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+			defer cancel()
+			return provider.Shutdown(shutdownCtx)
+		}
+	}
+
+	log.Infof(ctx, "Telemetry provider instance created successfully")
+	return provider, closer, nil
+}
+
+// Provider is the exported plugin instance
+var Provider = metricsProvider{}

pkg/plugin/implementation/otelsetup/cmd/plugin_test.go

@@ -0,0 +1,296 @@
+package main
+
+import (
+	"context"
+	"testing"
+
+	"github.com/beckn-one/beckn-onix/pkg/plugin/implementation/otelsetup"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestMetricsProviderNew_Success(t *testing.T) {
+	provider := metricsProvider{}
+
+	tests := []struct {
+		name   string
+		ctx    context.Context
+		config map[string]string
+	}{
+		{
+			name: "Valid config with all fields",
+			ctx:  context.Background(),
+			config: map[string]string{
+				"serviceName":    "test-service",
+				"serviceVersion": "1.0.0",
+				"enableMetrics":  "true",
+				"environment":    "test",
+			},
+		},
+		{
+			name:   "Valid config with minimal fields (uses defaults)",
+			ctx:    context.Background(),
+			config: map[string]string{},
+		},
+		{
+			name: "Valid config with enableMetrics false",
+			ctx:  context.Background(),
+			config: map[string]string{
+				"enableMetrics": "false",
+			},
+		},
+		{
+			name: "Valid config with partial fields",
+			ctx:  context.Background(),
+			config: map[string]string{
+				"serviceName":    "custom-service",
+				"serviceVersion": "2.0.0",
+			},
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			telemetryProvider, cleanup, err := provider.New(tt.ctx, tt.config)
+
+			require.NoError(t, err, "New() should not return error")
+			require.NotNil(t, telemetryProvider, "New() should return non-nil provider")
+
+			// Metrics server is started inside provider when enabled; MetricsHandler is not exposed.
+			if cleanup != nil {
+				err := cleanup()
+				assert.NoError(t, err, "cleanup() should not return error")
+			}
+		})
+	}
+}
+
+func TestMetricsProviderNew_Failure(t *testing.T) {
+	provider := metricsProvider{}
+
+	tests := []struct {
+		name    string
+		ctx     context.Context
+		config  map[string]string
+		wantErr bool
+	}{
+		{
+			name:    "Nil context",
+			ctx:     nil,
+			config:  map[string]string{},
+			wantErr: true,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			telemetryProvider, cleanup, err := provider.New(tt.ctx, tt.config)
+
+			if tt.wantErr {
+				assert.Error(t, err, "New() should return error for nil context")
+				assert.Nil(t, telemetryProvider, "New() should return nil provider on error")
+				assert.Nil(t, cleanup, "New() should return nil cleanup on error")
+			} else {
+				assert.NoError(t, err, "New() should not return error")
+				assert.NotNil(t, telemetryProvider, "New() should return non-nil provider")
+			}
+		})
+	}
+}
+
+func TestMetricsProviderNew_ConfigConversion(t *testing.T) {
+	provider := metricsProvider{}
+	ctx := context.Background()
+
+	tests := []struct {
+		name           string
+		config         map[string]string
+		expectedConfig *otelsetup.Config
+	}{
+		{
+			name: "All fields provided",
+			config: map[string]string{
+				"serviceName":    "my-service",
+				"serviceVersion": "3.0.0",
+				"enableMetrics":  "true",
+				"environment":    "production",
+			},
+			expectedConfig: &otelsetup.Config{
+				ServiceName:    "my-service",
+				ServiceVersion: "3.0.0",
+				EnableMetrics:  true,
+				Environment:    "production",
+			},
+		},
+		{
+			name:   "Empty config uses defaults",
+			config: map[string]string{},
+			expectedConfig: &otelsetup.Config{
+				ServiceName:    otelsetup.DefaultConfig().ServiceName,
+				ServiceVersion: otelsetup.DefaultConfig().ServiceVersion,
+				EnableMetrics:  true, // Default when not specified
+				Environment:    otelsetup.DefaultConfig().Environment,
+			},
+		},
+		{
+			name: "EnableMetrics false",
+			config: map[string]string{
+				"enableMetrics": "false",
+			},
+			expectedConfig: &otelsetup.Config{
+				ServiceName:    otelsetup.DefaultConfig().ServiceName,
+				ServiceVersion: otelsetup.DefaultConfig().ServiceVersion,
+				EnableMetrics:  false,
+				Environment:    otelsetup.DefaultConfig().Environment,
+			},
+		},
+		{
+			name: "Invalid enableMetrics defaults to true",
+			config: map[string]string{
+				"enableMetrics": "invalid",
+			},
+			expectedConfig: &otelsetup.Config{
+				ServiceName:    otelsetup.DefaultConfig().ServiceName,
+				ServiceVersion: otelsetup.DefaultConfig().ServiceVersion,
+				EnableMetrics:  true, // Defaults to true on parse error
+				Environment:    otelsetup.DefaultConfig().Environment,
+			},
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			telemetryProvider, cleanup, err := provider.New(ctx, tt.config)
+
+			require.NoError(t, err, "New() should not return error")
+			require.NotNil(t, telemetryProvider, "New() should return non-nil provider")
+
+			if cleanup != nil {
+				err := cleanup()
+				assert.NoError(t, err, "cleanup() should not return error")
+			}
+		})
+	}
+}
+
+func TestMetricsProviderNew_BooleanParsing(t *testing.T) {
+	provider := metricsProvider{}
+	ctx := context.Background()
+
+	tests := []struct {
+		name          string
+		enableMetrics string
+		expected      bool
+	}{
+		{
+			name:          "True string",
+			enableMetrics: "true",
+			expected:      true,
+		},
+		{
+			name:          "False string",
+			enableMetrics: "false",
+			expected:      false,
+		},
+		{
+			name:          "True uppercase",
+			enableMetrics: "TRUE",
+			expected:      true,
+		},
+		{
+			name:          "False uppercase",
+			enableMetrics: "FALSE",
+			expected:      false,
+		},
+		{
+			name:          "Invalid value defaults to true",
+			enableMetrics: "invalid",
+			expected:      true, // Defaults to true on parse error
+		},
+		{
+			name:          "Empty string defaults to true",
+			enableMetrics: "",
+			expected:      true, // Defaults to true when not specified
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			config := map[string]string{
+				"enableMetrics": tt.enableMetrics,
+			}
+
+			telemetryProvider, cleanup, err := provider.New(ctx, config)
+
+			require.NoError(t, err, "New() should not return error")
+			require.NotNil(t, telemetryProvider, "New() should return non-nil provider")
+
+			if cleanup != nil {
+				err := cleanup()
+				assert.NoError(t, err, "cleanup() should not return error")
+			}
+		})
+	}
+}
+
+func TestMetricsProviderNew_CleanupFunction(t *testing.T) {
+	provider := metricsProvider{}
+	ctx := context.Background()
+
+	config := map[string]string{
+		"serviceName":    "test-service",
+		"serviceVersion": "1.0.0",
+		"enableMetrics":  "true",
+		"environment":    "test",
+	}
+
+	telemetryProvider, cleanup, err := provider.New(ctx, config)
+
+	require.NoError(t, err, "New() should not return error")
+	require.NotNil(t, telemetryProvider, "New() should return non-nil provider")
+	require.NotNil(t, cleanup, "New() should return non-nil cleanup function")
+
+	// Test that cleanup can be called successfully
+	err = cleanup()
+	assert.NoError(t, err, "cleanup() should not return error")
+}
+
+func TestProviderVariable(t *testing.T) {
+	assert.NotNil(t, Provider, "Provider should not be nil")
+
+	// Verify Provider implements the interface correctly
+	ctx := context.Background()
+	config := map[string]string{
+		"serviceName":    "test",
+		"serviceVersion": "1.0.0",
+		"enableMetrics":  "true",
+	}
+
+	telemetryProvider, cleanup, err := Provider.New(ctx, config)
+
+	require.NoError(t, err, "Provider.New() should not return error")
+	require.NotNil(t, telemetryProvider, "Provider.New() should return non-nil provider")
+
+	if cleanup != nil {
+		err := cleanup()
+		assert.NoError(t, err, "cleanup() should not return error")
+	}
+}
+
+func TestMetricsProviderNew_DefaultValues(t *testing.T) {
+	provider := metricsProvider{}
+	ctx := context.Background()
+
+	// Test with completely empty config
+	config := map[string]string{}
+
+	telemetryProvider, cleanup, err := provider.New(ctx, config)
+
+	require.NoError(t, err, "New() should not return error with empty config")
+	require.NotNil(t, telemetryProvider, "New() should return non-nil provider")
+
+	if cleanup != nil {
+		err := cleanup()
+		assert.NoError(t, err, "cleanup() should not return error")
+	}
+}

pkg/plugin/implementation/otelsetup/otelsetup.go

@@ -0,0 +1,169 @@
+package otelsetup
+
+import (
+	"context"
+	"fmt"
+	"net"
+	"net/http"
+	"sync"
+	"time"
+
+	clientprom "github.com/prometheus/client_golang/prometheus"
+	clientpromhttp "github.com/prometheus/client_golang/prometheus/promhttp"
+	"go.opentelemetry.io/contrib/instrumentation/runtime"
+	"go.opentelemetry.io/otel"
+	"go.opentelemetry.io/otel/attribute"
+	otelprom "go.opentelemetry.io/otel/exporters/prometheus"
+	"go.opentelemetry.io/otel/sdk/metric"
+	"go.opentelemetry.io/otel/sdk/resource"
+
+	"github.com/beckn-one/beckn-onix/pkg/log"
+	"github.com/beckn-one/beckn-onix/pkg/plugin"
+	"github.com/beckn-one/beckn-onix/pkg/telemetry"
+)
+
+// Setup wires the telemetry provider. This is the concrete implementation
+// behind the OtelSetupMetricsProvider interface.
+type Setup struct{}
+
+// Config represents OpenTelemetry related configuration.
+type Config struct {
+	ServiceName    string `yaml:"serviceName"`
+	ServiceVersion string `yaml:"serviceVersion"`
+	EnableMetrics  bool   `yaml:"enableMetrics"`
+	Environment    string `yaml:"environment"`
+	MetricsPort    string `yaml:"metricsPort"`
+}
+
+// DefaultConfig returns sensible defaults for telemetry configuration.
+func DefaultConfig() *Config {
+	return &Config{
+		ServiceName:    "beckn-onix",
+		ServiceVersion: "dev",
+		EnableMetrics:  true,
+		Environment:    "development",
+		MetricsPort:    "9090",
+	}
+}
+
+// ToPluginConfig converts Config to plugin.Config format.
+func ToPluginConfig(cfg *Config) *plugin.Config {
+	return &plugin.Config{
+		ID: "otelsetup",
+		Config: map[string]string{
+			"serviceName":    cfg.ServiceName,
+			"serviceVersion": cfg.ServiceVersion,
+			"enableMetrics":  fmt.Sprintf("%t", cfg.EnableMetrics),
+			"environment":    cfg.Environment,
+			"metricsPort":    cfg.MetricsPort,
+		},
+	}
+}
+
+// New initializes the underlying telemetry provider. The returned provider
+// exposes the HTTP handler and shutdown hooks that the core application can
+// manage directly.
+func (Setup) New(ctx context.Context, cfg *Config) (*telemetry.Provider, error) {
+	if cfg == nil {
+		return nil, fmt.Errorf("telemetry config cannot be nil")
+	}
+
+	// Apply defaults if fields are empty
+	if cfg.ServiceName == "" {
+		cfg.ServiceName = DefaultConfig().ServiceName
+	}
+	if cfg.ServiceVersion == "" {
+		cfg.ServiceVersion = DefaultConfig().ServiceVersion
+	}
+	if cfg.Environment == "" {
+		cfg.Environment = DefaultConfig().Environment
+	}
+	if cfg.MetricsPort == "" {
+		cfg.MetricsPort = DefaultConfig().MetricsPort
+	}
+
+	if !cfg.EnableMetrics {
+		log.Info(ctx, "OpenTelemetry metrics disabled")
+		return &telemetry.Provider{
+			Shutdown: func(context.Context) error { return nil },
+		}, nil
+	}
+
+	res, err := resource.New(
+		ctx,
+		resource.WithAttributes(
+			attribute.String("service.name", cfg.ServiceName),
+			attribute.String("service.version", cfg.ServiceVersion),
+			attribute.String("deployment.environment", cfg.Environment),
+		),
+	)
+	if err != nil {
+		return nil, fmt.Errorf("failed to create telemetry resource: %w", err)
+	}
+
+	registry := clientprom.NewRegistry()
+
+	exporter, err := otelprom.New(
+		otelprom.WithRegisterer(registry),
+		otelprom.WithoutUnits(),
+		otelprom.WithoutScopeInfo(),
+	)
+	if err != nil {
+		return nil, fmt.Errorf("failed to create prometheus exporter: %w", err)
+	}
+
+	meterProvider := metric.NewMeterProvider(
+		metric.WithReader(exporter),
+		metric.WithResource(res),
+	)
+
+	otel.SetMeterProvider(meterProvider)
+	log.Infof(ctx, "OpenTelemetry metrics initialized for service=%s version=%s env=%s",
+		cfg.ServiceName, cfg.ServiceVersion, cfg.Environment)
+
+	if err := runtime.Start(runtime.WithMinimumReadMemStatsInterval(0)); err != nil {
+		log.Warnf(ctx, "Failed to start Go runtime instrumentation: %v", err)
+	}
+
+	// Create metrics handler
+	metricsHandler := clientpromhttp.HandlerFor(registry, clientpromhttp.HandlerOpts{})
+
+	// Create and start metrics HTTP server
+	metricsMux := http.NewServeMux()
+	metricsMux.Handle("/metrics", metricsHandler)
+
+	metricsServer := &http.Server{
+		Addr:         net.JoinHostPort("", cfg.MetricsPort),
+		Handler:      metricsMux,
+		ReadTimeout:  10 * time.Second,
+		WriteTimeout: 10 * time.Second,
+		IdleTimeout:  30 * time.Second,
+	}
+
+	var serverWg sync.WaitGroup
+	serverWg.Add(1)
+	go func() {
+		defer serverWg.Done()
+		log.Infof(ctx, "Metrics server listening on %s", metricsServer.Addr)
+		if err := metricsServer.ListenAndServe(); err != nil && err != http.ErrServerClosed {
+			log.Errorf(ctx, fmt.Errorf("metrics server ListenAndServe: %w", err), "error listening and serving metrics")
+		}
+	}()
+
+	return &telemetry.Provider{
+		MeterProvider:  meterProvider,
+		MetricsHandler: metricsHandler,
+		Shutdown: func(shutdownCtx context.Context) error {
+			log.Infof(ctx, "Shutting down metrics server...")
+			// Shutdown the metrics server
+			serverShutdownCtx, cancel := context.WithTimeout(shutdownCtx, 10*time.Second)
+			defer cancel()
+			if err := metricsServer.Shutdown(serverShutdownCtx); err != nil {
+				log.Errorf(ctx, fmt.Errorf("metrics server shutdown: %w", err), "error shutting down metrics server")
+			}
+			serverWg.Wait()
+			// Shutdown the meter provider
+			return meterProvider.Shutdown(shutdownCtx)
+		},
+	}, nil
+}

pkg/plugin/implementation/otelsetup/otelsetup_test.go

@@ -0,0 +1,259 @@
+package otelsetup
+
+import (
+	"context"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestSetup_New_Success(t *testing.T) {
+	setup := Setup{}
+	ctx := context.Background()
+
+	tests := []struct {
+		name string
+		cfg  *Config
+	}{
+		{
+			name: "Valid config with all fields",
+			cfg: &Config{
+				ServiceName:    "test-service",
+				ServiceVersion: "1.0.0",
+				EnableMetrics:  true,
+				Environment:    "test",
+			},
+		},
+		{
+			name: "Valid config with metrics disabled",
+			cfg: &Config{
+				ServiceName:    "test-service",
+				ServiceVersion: "1.0.0",
+				EnableMetrics:  false,
+				Environment:    "test",
+			},
+		},
+		{
+			name: "Config with empty fields uses defaults",
+			cfg: &Config{
+				ServiceName:    "",
+				ServiceVersion: "",
+				EnableMetrics:  true,
+				Environment:    "",
+			},
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			provider, err := setup.New(ctx, tt.cfg)
+
+			require.NoError(t, err, "New() should not return error")
+			require.NotNil(t, provider, "New() should return non-nil provider")
+			require.NotNil(t, provider.Shutdown, "Provider should have shutdown function")
+
+			if tt.cfg.EnableMetrics {
+				assert.NotNil(t, provider.MeterProvider, "MeterProvider should be set when metrics enabled")
+			}
+
+			// Test shutdown
+			err = provider.Shutdown(ctx)
+			assert.NoError(t, err, "Shutdown should not return error")
+		})
+	}
+}
+
+func TestSetup_New_Failure(t *testing.T) {
+	setup := Setup{}
+	ctx := context.Background()
+
+	tests := []struct {
+		name    string
+		cfg     *Config
+		wantErr bool
+	}{
+		{
+			name:    "Nil config",
+			cfg:     nil,
+			wantErr: true,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			provider, err := setup.New(ctx, tt.cfg)
+
+			if tt.wantErr {
+				assert.Error(t, err, "New() should return error")
+				assert.Nil(t, provider, "New() should return nil provider on error")
+			} else {
+				assert.NoError(t, err, "New() should not return error")
+				assert.NotNil(t, provider, "New() should return non-nil provider")
+			}
+		})
+	}
+}
+
+func TestSetup_New_DefaultValues(t *testing.T) {
+	setup := Setup{}
+	ctx := context.Background()
+
+	// Test with empty fields - should use defaults
+	cfg := &Config{
+		ServiceName:    "",
+		ServiceVersion: "",
+		EnableMetrics:  true,
+		Environment:    "",
+	}
+
+	provider, err := setup.New(ctx, cfg)
+	require.NoError(t, err)
+	require.NotNil(t, provider)
+
+	// Verify defaults are applied by checking that provider is functional
+	assert.NotNil(t, provider.MeterProvider, "MeterProvider should be set with defaults")
+
+	// Cleanup
+	err = provider.Shutdown(ctx)
+	assert.NoError(t, err)
+}
+
+func TestSetup_New_MetricsDisabled(t *testing.T) {
+	setup := Setup{}
+	ctx := context.Background()
+
+	cfg := &Config{
+		ServiceName:    "test-service",
+		ServiceVersion: "1.0.0",
+		EnableMetrics:  false,
+		Environment:    "test",
+	}
+
+	provider, err := setup.New(ctx, cfg)
+	require.NoError(t, err)
+	require.NotNil(t, provider)
+
+	// When metrics are disabled, MetricsHandler should be nil and MeterProvider should be nil
+	assert.Nil(t, provider.MeterProvider, "MeterProvider should be nil when metrics disabled")
+
+	// Shutdown should still work
+	err = provider.Shutdown(ctx)
+	assert.NoError(t, err, "Shutdown should work even when metrics disabled")
+}
+
+func TestToPluginConfig_Success(t *testing.T) {
+	tests := []struct {
+		name           string
+		cfg            *Config
+		expectedID     string
+		expectedConfig map[string]string
+	}{
+		{
+			name: "Valid config with all fields",
+			cfg: &Config{
+				ServiceName:    "test-service",
+				ServiceVersion: "1.0.0",
+				EnableMetrics:  true,
+				Environment:    "test",
+			},
+			expectedID: "otelsetup",
+			expectedConfig: map[string]string{
+				"serviceName":    "test-service",
+				"serviceVersion": "1.0.0",
+				"enableMetrics":  "true",
+				"environment":    "test",
+				"metricsPort":    "",
+			},
+		},
+		{
+			name: "Config with enableMetrics false",
+			cfg: &Config{
+				ServiceName:    "my-service",
+				ServiceVersion: "2.0.0",
+				EnableMetrics:  false,
+				Environment:    "production",
+			},
+			expectedID: "otelsetup",
+			expectedConfig: map[string]string{
+				"serviceName":    "my-service",
+				"serviceVersion": "2.0.0",
+				"enableMetrics":  "false",
+				"environment":    "production",
+				"metricsPort":    "",
+			},
+		},
+		{
+			name: "Config with empty fields",
+			cfg: &Config{
+				ServiceName:    "",
+				ServiceVersion: "",
+				EnableMetrics:  true,
+				Environment:    "",
+			},
+			expectedID: "otelsetup",
+			expectedConfig: map[string]string{
+				"serviceName":    "",
+				"serviceVersion": "",
+				"enableMetrics":  "true",
+				"environment":    "",
+				"metricsPort":    "",
+			},
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := ToPluginConfig(tt.cfg)
+
+			require.NotNil(t, result, "ToPluginConfig should return non-nil config")
+			assert.Equal(t, tt.expectedID, result.ID, "Plugin ID should be 'otelsetup'")
+			assert.Equal(t, tt.expectedConfig, result.Config, "Config map should match expected values")
+		})
+	}
+}
+
+func TestToPluginConfig_NilConfig(t *testing.T) {
+	// Test that ToPluginConfig handles nil config
+	// Note: This will panic if nil is passed, which is acceptable behavior
+	// as the function expects a valid config. In practice, callers should check for nil.
+	assert.Panics(t, func() {
+		ToPluginConfig(nil)
+	}, "ToPluginConfig should panic when given nil config")
+}
+
+func TestToPluginConfig_BooleanConversion(t *testing.T) {
+	tests := []struct {
+		name          string
+		enableMetrics bool
+		expected      string
+	}{
+		{
+			name:          "EnableMetrics true",
+			enableMetrics: true,
+			expected:      "true",
+		},
+		{
+			name:          "EnableMetrics false",
+			enableMetrics: false,
+			expected:      "false",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			cfg := &Config{
+				ServiceName:    "test",
+				ServiceVersion: "1.0.0",
+				EnableMetrics:  tt.enableMetrics,
+				Environment:    "test",
+				MetricsPort:    "",
+			}
+
+			result := ToPluginConfig(cfg)
+			require.NotNil(t, result)
+			assert.Equal(t, tt.expected, result.Config["enableMetrics"], "enableMetrics should be converted to string correctly")
+			assert.Equal(t, "", result.Config["metricsPort"], "metricsPort should be included even when empty")
+		})
+	}
+}

pkg/plugin/implementation/router/router_test.go

@@ -683,7 +683,6 @@ func TestExcludeActionWithNonURLTargetTypes(t *testing.T) {
 	}
 }
 
-
 // TestV2RouteSuccess tests v2 routing with domain-agnostic behavior
 func TestV2RouteSuccess(t *testing.T) {
 	ctx := context.Background()

pkg/plugin/manager.go

@@ -15,6 +15,7 @@ import (
 
 	"github.com/beckn-one/beckn-onix/pkg/log"
 	"github.com/beckn-one/beckn-onix/pkg/plugin/definition"
+	"github.com/beckn-one/beckn-onix/pkg/telemetry"
 )
 
 type onixPlugin interface {
@@ -196,6 +197,33 @@ func (m *Manager) Middleware(ctx context.Context, cfg *Config) (func(http.Handle
 	return mwp.New(ctx, cfg.Config)
 }
 
+// OtelSetup initializes OpenTelemetry via a dedicated plugin. The plugin is
+// expected to return a telemetry Provider that the core application can use for
+// instrumentation.
+func (m *Manager) OtelSetup(ctx context.Context, cfg *Config) (*telemetry.Provider, error) {
+	if cfg == nil {
+		log.Info(ctx, "Telemetry config not provided; skipping OpenTelemetry setup")
+		return nil, nil
+	}
+
+	otp, err := provider[definition.OtelSetupMetricsProvider](m.plugins, cfg.ID)
+	if err != nil {
+		return nil, fmt.Errorf("failed to load provider for %s: %w", cfg.ID, err)
+	}
+	provider, closer, err := otp.New(ctx, cfg.Config)
+	if err != nil {
+		return nil, err
+	}
+	if closer != nil {
+		m.closers = append(m.closers, func() {
+			if err := closer(); err != nil {
+				log.Errorf(context.Background(), err, "Failed to shutdown telemetry provider")
+			}
+		})
+	}
+	return provider, nil
+}
+
 // TransportWrapper returns a TransportWrapper instance based on the provided configuration.
 func (m *Manager) TransportWrapper(ctx context.Context, cfg *Config) (definition.TransportWrapper, error) {
 	twp, err := provider[definition.TransportWrapperProvider](m.plugins, cfg.ID)