docker-compose.observability-dev.yml

# =============================================================================
# OpenTelemetry Development Stack (TASK-034: Observability Foundation)
# =============================================================================
# Replaces legacy 10-container monitoring stack with single grafana/otel-lgtm
# container for unified metrics, logs, traces collection.
#
# Components included in grafana/otel-lgtm:
#   - Grafana (UI) - port 3000
#   - Loki (logs)
#   - Tempo (traces)
#   - Mimir (metrics)
#   - OpenTelemetry Collector (OTLP receiver)
#
# OTLP Endpoints:
#   - gRPC: port 4317
#   - HTTP: port 4318
#
# Usage:
#   make otel-start    # Start observability stack
#   make otel-stop     # Stop observability stack
#   make otel-logs     # View container logs
#
# Related: ADR-014-observability-stack-modernization.md
# =============================================================================

services:

  otel-lgtm:
    hostname: ${APP_OTEL_LGTM_NAME:-test-otel-lgtm}
    container_name: ${APP_OTEL_LGTM_NAME:-test-otel-lgtm}
    image: grafana/otel-lgtm:latest
    restart: unless-stopped
    ports:
      # Grafana UI
      - "${APP_GRAFANA_EXT_PORT:-3000}:${APP_GRAFANA_PORT:-3000}"
      # OTLP gRPC receiver
      - "${APP_OTLP_GRPC_EXT_PORT:-4317}:${APP_OTLP_GRPC_PORT:-4317}"
      # OTLP HTTP receiver
      - "${APP_OTLP_HTTP_EXT_PORT:-4318}:${APP_OTLP_HTTP_PORT:-4318}"
    environment:
      # Grafana configuration
      - GF_SECURITY_ADMIN_USER=${APP_GRAFANA_ADMIN_USER:-admin}
      - GF_SECURITY_ADMIN_PASSWORD=${APP_GRAFANA_ADMIN_PASSWORD:-admin}
      - GF_USERS_ALLOW_SIGN_UP=false
      # Enable anonymous access for development
      - GF_AUTH_ANONYMOUS_ENABLED=true
      - GF_AUTH_ANONYMOUS_ORG_ROLE=Admin
      # Logging level (debug for provisioning diagnostics)
      - GF_LOG_LEVEL=debug
      # Enable Grafana logging in otel-lgtm container (default is false)
      - ENABLE_LOGS_GRAFANA=true
      # Enable OTel Collector logging for debugging
      - ENABLE_LOGS_OTELCOL=true
      # Enable unified alerting for file-based provisioning (TASK-037)
      - GF_UNIFIED_ALERTING_ENABLED=true
      - GF_ALERTING_ENABLED=false
      # CRITICAL: otel-lgtm sets GF_PATHS_HOME=/data/grafana, so we must explicitly
      # set provisioning path to where our files are mounted
      - GF_PATHS_PROVISIONING=/etc/grafana/provisioning
    volumes:
      # Persist Grafana dashboards and settings
      - otel_grafana_data:/var/lib/grafana
      # Persist Loki data (logs)
      - otel_loki_data:/loki
      # Persist Tempo data (traces)
      - otel_tempo_data:/tmp/tempo
      # Persist Mimir data (metrics)
      - otel_mimir_data:/data/mimir
      # Prometheus scrape configuration (TASK-037: Enables infrastructure metrics)
      - ./.docker/grafana/prometheus.yml:/otel-lgtm/prometheus.yaml:ro
      # Grafana provisioning (TASK-037: Alerting & Profiling)
      - ./.docker/grafana/provisioning/dashboards:/etc/grafana/provisioning/dashboards:ro
      - ./.docker/grafana/provisioning/datasources:/etc/grafana/provisioning/datasources:ro
      - ./.docker/grafana/provisioning/alerting:/etc/grafana/provisioning/alerting:ro
    networks:
      - ${DOCKER_NETWORK_NAME:-test_article_net}
    healthcheck:
      test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider http://localhost:3000/api/health || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 60s
    labels:
      - "org.label-schema.group=observability"
      - "org.label-schema.description=OpenTelemetry LGTM Stack (Loki, Grafana, Tempo, Mimir)"
    deploy:
      resources:
        limits:
          memory: 1G
        reservations:
          memory: 512M

  # =============================================================================
  # Infrastructure Exporters (TASK-036: Infrastructure Integration)
  # =============================================================================

  # PostgreSQL Exporter - Exposes PostgreSQL metrics for Prometheus/Mimir
  # Metrics endpoint: http://localhost:9187/metrics
  postgres-exporter:
    image: quay.io/prometheuscommunity/postgres-exporter:latest
    container_name: ${CI_SERVICE_NAME:-test-micro-article-system}-postgres-exporter
    hostname: postgres-exporter
    restart: unless-stopped
    environment:
      DATA_SOURCE_NAME: "postgresql://${APP_DATABASE_SERVICE_LOGIN:-article_service}:${APP_DATABASE_SERVICE_PASSWORD:-article_service_pwd}@${APP_DATABASE_HOST:-test-micro-article-system-database}:${APP_DATABASE_PORT:-5432}/${APP_DATABASE_NAME:-article}?sslmode=disable"
    ports:
      - "${POSTGRES_EXPORTER_PORT:-9187}:9187"
    networks:
      - ${DOCKER_NETWORK_NAME:-test_article_net}
    # Note: Depends on PostgreSQL from main docker-compose.yml
    # Start main stack first: make start
    labels:
      - "org.label-schema.group=observability"
      - "org.label-schema.description=PostgreSQL Prometheus Exporter"
    deploy:
      resources:
        limits:
          memory: 128M
        reservations:
          memory: 64M

  # Redis Exporter - Exposes Redis metrics for Prometheus/Mimir
  # Metrics endpoint: http://localhost:9121/metrics
  redis-exporter:
    image: oliver006/redis_exporter:latest
    container_name: ${CI_SERVICE_NAME:-test-micro-article-system}-redis-exporter
    hostname: redis-exporter
    restart: unless-stopped
    environment:
      REDIS_ADDR: "redis://${APP_REDIS_HOST:-test-micro-article-system-redis}:${APP_REDIS_PORT:-6379}"
    ports:
      - "${REDIS_EXPORTER_PORT:-9121}:9121"
    networks:
      - ${DOCKER_NETWORK_NAME:-test_article_net}
    # Note: Depends on Redis from main docker-compose.yml
    # Start main stack first: make start
    labels:
      - "org.label-schema.group=observability"
      - "org.label-schema.description=Redis Prometheus Exporter"
    deploy:
      resources:
        limits:
          memory: 64M
        reservations:
          memory: 32M

  # =============================================================================
  # Webhook Receiver for Alert Testing (TASK-037: Alerting & Profiling)
  # =============================================================================

  # Simple HTTP echo server to receive and log alert notifications
  # Logs all incoming webhooks for debugging/testing alert delivery
  # View logs: docker logs -f test-micro-article-system-webhook-receiver
  webhook-receiver:
    image: mendhak/http-https-echo:latest
    container_name: ${CI_SERVICE_NAME:-test-micro-article-system}-webhook-receiver
    hostname: webhook-receiver
    restart: unless-stopped
    environment:
      - HTTP_PORT=9999
      - HTTPS_PORT=9443
      # Log full body for debugging
      - LOG_WITHOUT_NEWLINE=true
    ports:
      - "${WEBHOOK_RECEIVER_PORT:-9999}:9999"
    networks:
      - ${DOCKER_NETWORK_NAME:-test_article_net}
    labels:
      - "org.label-schema.group=observability"
      - "org.label-schema.description=Webhook receiver for alert testing"
    deploy:
      resources:
        limits:
          memory: 64M
        reservations:
          memory: 32M

  # =============================================================================
  # Pyroscope Continuous Profiling (TASK-037: Alerting & Profiling)
  # =============================================================================

  # Pyroscope - Continuous profiling for performance analysis
  # UI: http://localhost:4040
  # Integrated with Grafana as data source for unified observability
  pyroscope:
    image: grafana/pyroscope:latest
    container_name: ${CI_SERVICE_NAME:-test-micro-article-system}-pyroscope
    hostname: pyroscope
    restart: unless-stopped
    # Run in all-in-one mode (single-node) - explicitly set for clarity
    # Uses filesystem storage for development environments
    command:
      - "-target=all"
      - "-self-profiling.disable-push=true"
      - "-memberlist.cluster-label="
      - "-memberlist.join="
    ports:
      - "${PYROSCOPE_PORT:-4040}:4040"
    environment:
      # Disable self-profiling to reduce overhead in development
      - PYROSCOPE_SELF_PROFILING_DISABLE=true
    volumes:
      - pyroscope_data:/data
    networks:
      - ${DOCKER_NETWORK_NAME:-test_article_net}
    healthcheck:
      test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider http://localhost:4040/ready || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 30s
    labels:
      - "org.label-schema.group=observability"
      - "org.label-schema.description=Pyroscope Continuous Profiling"
    deploy:
      resources:
        limits:
          memory: 512M
        reservations:
          memory: 256M

networks:
  test_article_net:
    external: true
    name: ${DOCKER_NETWORK_NAME:-test_article_net}

volumes:
  otel_grafana_data:
    name: ${APP_COMPANY_NAME:-test}_otel_grafana_data
  otel_loki_data:
    name: ${APP_COMPANY_NAME:-test}_otel_loki_data
  otel_tempo_data:
    name: ${APP_COMPANY_NAME:-test}_otel_tempo_data
  otel_mimir_data:
    name: ${APP_COMPANY_NAME:-test}_otel_mimir_data
  pyroscope_data:
    name: ${APP_COMPANY_NAME:-test}_pyroscope_data