✨ Add new dev-run-compose task to Taskfile for running the CAATSM receiver in a Docker Compose development stack. Update monitoring configuration in config.dev.toml to use console format. Modify Prometheus scrape configuration to target a specific IP address. Enhance NATS consumer error handling with retry logic for JetStream availability and adjust logging level for consumer metrics.

Taskfile.yml

@@ -147,6 +147,13 @@ tasks:
       - echo "Stopping dev infrastructure..."
       - docker compose -f docker-compose.dev.yml down -v
 
+  dev-run-compose:
+    desc: Run receiver inside docker-compose dev stack
+    cmds:
+      - echo "Starting dev stack including caatsm-receiver..."
+      - docker compose -f docker-compose.dev.yml up -d postgres nats nats-box nats-exporter
+      - docker compose -f docker-compose.dev.yml up -d otel-collector jaeger prometheus grafana caatsm-receiver
+
   dev-run:
     desc: Run receiver locally against dev stack
     deps:
@@ -158,6 +165,9 @@ tasks:
       CAATSM_TELEMETRY_ENABLED: "true"
       CAATSM_TELEMETRY_ENDPOINT: localhost:4318
       CAATSM_TELEMETRY_INSECURE: "true"
+      CAATSM_MONITORING_ADDR: 0.0.0.0:2112
+      CAATSM_MONITORING_ENABLE_METRICS: "true"
+      CAATSM_MONITORING_ENABLE_HEALTH: "true"
       GO_ENV: dev
     cmds:
       - |
@@ -174,9 +184,9 @@ tasks:
         GO_ENV=dev \
         NATS_URL=${CAATSM_NATS_URL:-nats://localhost:4222} \
         SUBJECT=${CAATSM_NATS_SUBJECT:-telegram.serial} \
-        COUNT=${COUNT:-10} \
-        CATEGORY=${CATEGORY:-mixed} \
-        STATUS=${STATUS:-random} \
+        COUNT={{.COUNT | default "10"}} \
+        CATEGORY={{.CATEGORY | default "mixed"}} \
+        STATUS={{.STATUS | default "random"}} \
         bash -c '
           set -euo pipefail
           cmd=(go run ./cmd/seed-telegrams)

configs/config.dev.toml

@@ -49,7 +49,7 @@ monitor_interval = "30s"
 
 [log]
 level = "info"
-format = "json"
+format = "console"
 
 [telemetry]
 enabled = true

configs/prometheus.dev.yml

@@ -15,7 +15,5 @@ scrape_configs:
   - job_name: "caatsm-receiver"
     static_configs:
       - targets:
-          # When go-caatsm runs on the host/WSL and Prometheus runs in Docker,
-          # use host.docker.internal to reach the host monitoring server.
-          - "host.docker.internal:2112"
+        - 172.23.189.12:2112
 

internal/infra/nats/consumer.go

@@ -241,6 +241,17 @@ func (c *Consumer) startJetStream(ctx context.Context) error {
 				// Timeout is expected when no messages are available
 				continue
 			}
+			if errors.Is(err, nats.ErrNoResponders) {
+				// JetStream API is currently unavailable (e.g., NATS just restarted or JetStream not ready).
+				// Back off a bit to avoid log spam while allowing the system to recover.
+				c.logger.Warn("JetStream not available, will retry",
+					zap.Error(err),
+					zap.String("stream", c.streamName),
+					zap.String("consumer", c.consumerName),
+				)
+				time.Sleep(5 * time.Second)
+				continue
+			}
 			c.logger.Error("Failed to fetch messages", zap.Error(err))
 			time.Sleep(time.Second)
 			continue
@@ -377,7 +388,7 @@ func (c *Consumer) emitConsumerStats(ctx context.Context) {
 				continue
 			}
 
-			c.logger.Info("JetStream consumer metrics",
+			c.logger.Debug("JetStream consumer metrics",
 				zap.String("stream", c.streamName),
 				zap.String("consumer", c.consumerName),
 				zap.Uint64("num_ack_pending", uint64(info.NumAckPending)),

internal/infra/nats/jetstream.go

@@ -18,6 +18,8 @@ func ProvideNATSConn(cfg *config.Config, logger *zap.Logger) (*nats.Conn, error)
 		nats.RetryOnFailedConnect(true),
 		nats.Timeout(cfg.Timeouts.Server),
 		nats.ReconnectWait(cfg.Timeouts.ReconnectWait),
+		// Use infinite reconnects so the app survives long NATS outages (e.g. docker compose down/up).
+		nats.MaxReconnects(-1),
 		nats.DisconnectErrHandler(func(nc *nats.Conn, err error) {
 			if err != nil {
 				logger.Warn("NATS disconnected", zap.Error(err))