feat(monitoring): integrate prometheus and promtail log aggregation with sms alerts

docker/docker-compose.staging.template.yaml

@@ -108,6 +108,37 @@ services:
     networks:
       - baron_net
 
+  grafana-sms-webhook:
+    image: ${SMS_WEBHOOK_IMAGE_NAME:-ai-team/grafana-sms-webhook}:${IMAGE_TAG:-latest}
+    container_name: grafana_sms_webhook
+    restart: unless-stopped
+    env_file:
+      - .env
+    environment:
+      - NAVER_CLOUD_ACCESS_KEY=${NAVER_CLOUD_ACCESS_KEY}
+      - NAVER_CLOUD_SECRET_KEY=${NAVER_CLOUD_SECRET_KEY}
+      - NAVER_CLOUD_SERVICE_ID=${NAVER_CLOUD_SERVICE_ID}
+      - NAVER_SENDER_PHONE_NUMBER=${NAVER_SENDER_PHONE_NUMBER}
+      - MONITOR_RECIPIENT_PHONES=${MONITOR_RECIPIENT_PHONES}
+    ports:
+      - "${SMS_WEBHOOK_PORT:-8080}:8080"
+    networks:
+      - baron_net
+
+  promtail:
+    image: grafana/promtail:2.9.0
+    container_name: baron_promtail
+    restart: unless-stopped
+    volumes:
+      - /var/run/docker.sock:/var/run/docker.sock:ro
+      - /var/lib/docker/containers:/var/lib/docker/containers:ro
+      - ./promtail-config.staging.template.yaml:/etc/promtail/promtail-config.yaml:ro
+    command: -config.file=/etc/promtail/promtail-config.yaml
+    environment:
+      - LOKI_URL=${LOKI_URL:-http://loki:3100/loki/api/v1/push}
+    networks:
+      - baron_net
+
 networks:
   baron_net:
     external: true

docker/promtail-config.staging.template.yaml

@@ -0,0 +1,33 @@
+server:
+  http_listen_port: 9080
+  grpc_listen_port: 0
+
+positions:
+  filename: /tmp/positions.yaml
+
+clients:
+  - url: ${LOKI_URL:-http://loki:3100/loki/api/v1/push}
+
+scrape_configs:
+  - job_name: baron-sso-container-logs
+    docker_sd_configs:
+      - host: unix:///var/run/docker.sock
+        refresh_interval: 10s
+    relabel_configs:
+      - source_labels: ['__meta_docker_container_name']
+        regex: '/(.*)'
+        target_label: 'container_name'
+      # Baron SSO 및 핵심 Ory Stack 컨테이너만 필터링하여 로그 수집
+      - source_labels: ['container_name']
+        regex: '(baron_.*|oathkeeper|kratos|hydra|keto)'
+        action: keep
+      # 컨테이너 명에서 앞의 접두사를 떼어 서비스 및 잡 라벨 부여 (예: baron_backend -> backend)
+      - source_labels: ['container_name']
+        regex: 'baron_(.*)'
+        target_label: 'service'
+      - source_labels: ['container_name']
+        regex: 'baron_(.*)'
+        target_label: 'job'
+      # 정적 라벨 추가
+      - target_label: 'app_env'
+        replacement: 'stage'