Добавлены метрики Prometheus и дашборды Grafana (этап 2.5)

Все три Go-сервиса (ingest, device-control, rule-engine) отдают
/metrics в формате Prometheus: счётчики MQTT/RabbitMQ/ClickHouse
операций, гистограммы длительности батч-флашей и диспатча правил,
переходы устройств online/offline. Добавлены сервисы prometheus и
grafana в docker-compose с провижининг конфигом (datasource +
два готовых дашборда: "Ingest & Telemetry" и "Automation & Devices").

services/ingest-service/internal/config/config.go

@@ -24,6 +24,8 @@ type Config struct {
 	BatchMaxSize       int
 	BatchFlushInterval time.Duration
 	BatchFlushTimeout  time.Duration
+
+	MetricsPort int
 }
 
 func Load() (Config, error) {
@@ -55,6 +57,9 @@ func Load() (Config, error) {
 	if cfg.BatchFlushTimeout, err = getEnvDuration("INGEST_BATCH_FLUSH_TIMEOUT", 10*time.Second); err != nil {
 		return Config{}, err
 	}
+	if cfg.MetricsPort, err = getEnvInt("INGEST_METRICS_PORT", 9101); err != nil {
+		return Config{}, err
+	}
 
 	return cfg, nil
 }

services/ingest-service/internal/metrics/metrics.go

@@ -0,0 +1,48 @@
+// Package metrics defines ingest-service's Prometheus metrics. Registered
+// automatically (via promauto) into the default registry on import; served
+// by main.go's dedicated metrics HTTP server.
+package metrics
+
+import (
+	"github.com/prometheus/client_golang/prometheus"
+	"github.com/prometheus/client_golang/prometheus/promauto"
+)
+
+var (
+	MQTTMessagesReceived = promauto.NewCounter(prometheus.CounterOpts{
+		Name: "ingest_mqtt_messages_received_total",
+		Help: "Total number of telemetry messages received from MQTT.",
+	})
+
+	MQTTMessagesInvalid = promauto.NewCounter(prometheus.CounterOpts{
+		Name: "ingest_mqtt_messages_invalid_total",
+		Help: "Total number of telemetry messages dropped for failing validation.",
+	})
+
+	MQTTMessagesDropped = promauto.NewCounter(prometheus.CounterOpts{
+		Name: "ingest_mqtt_messages_dropped_total",
+		Help: "Total number of readings dropped because the worker backlog was full.",
+	})
+
+	BatchFlushes = promauto.NewCounterVec(prometheus.CounterOpts{
+		Name: "ingest_batch_flush_total",
+		Help: "Total number of batch flushes to ClickHouse, by outcome.",
+	}, []string{"outcome"})
+
+	BatchFlushDuration = promauto.NewHistogram(prometheus.HistogramOpts{
+		Name:    "ingest_batch_flush_duration_seconds",
+		Help:    "Duration of ClickHouse batch insert calls.",
+		Buckets: prometheus.DefBuckets,
+	})
+
+	BatchSize = promauto.NewHistogram(prometheus.HistogramOpts{
+		Name:    "ingest_batch_size",
+		Help:    "Number of readings per flushed batch.",
+		Buckets: []float64{1, 5, 10, 25, 50, 100, 250, 500, 1000},
+	})
+
+	RabbitMQPublishes = promauto.NewCounterVec(prometheus.CounterOpts{
+		Name: "ingest_rabbitmq_publish_total",
+		Help: "Total number of new-reading events published to RabbitMQ, by outcome.",
+	}, []string{"outcome"})
+)