docker-stack.yml

name: latitude

x-logging: &logging
  driver: json-file
  options:
    max-size: "10m"
    max-file: "5"

# Each application service waits for the one-shot `migrations` job to finish
x-app: &app
  restart: unless-stopped
  env_file:
    - .env.production
  logging: *logging
  depends_on:
    migrations:
      condition: service_completed_successfully

services:
  # ==========================================================================
  # Application services (published images: latitudedata/<service>)
  # ==========================================================================
  web:
    <<: *app
    image: latitudedata/web:${LAT_IMAGE_TAG:-latest}
    ports:
      - "3000:3000"
    healthcheck:
      test:
        [
          "CMD-SHELL",
          "curl -fsS http://localhost:$${LAT_WEB_PORT:-3000}/api/health || exit 1",
        ]
      interval: 15s
      timeout: 5s
      retries: 10
      start_period: 40s
    deploy:
      replicas: 1
      restart_policy:
        condition: any
        delay: 5s

  api:
    <<: *app
    image: latitudedata/api:${LAT_IMAGE_TAG:-latest}
    ports:
      - "3001:3001"
    healthcheck:
      test:
        [
          "CMD-SHELL",
          "curl -fsS http://localhost:$${LAT_API_PORT:-3001}/health || exit 1",
        ]
      interval: 15s
      timeout: 5s
      retries: 10
      start_period: 40s
    deploy:
      replicas: 1
      restart_policy:
        condition: any
        delay: 5s

  ingest:
    <<: *app
    image: latitudedata/ingest:${LAT_IMAGE_TAG:-latest}
    ports:
      - "3002:3002"
    healthcheck:
      test:
        [
          "CMD-SHELL",
          "curl -fsS http://localhost:$${LAT_INGEST_PORT:-3002}/health || exit 1",
        ]
      interval: 15s
      timeout: 5s
      retries: 10
      start_period: 40s
    deploy:
      replicas: 1
      restart_policy:
        condition: any
        delay: 5s

  workers:
    <<: *app
    image: latitudedata/workers:${LAT_IMAGE_TAG:-latest}
    healthcheck:
      test:
        [
          "CMD-SHELL",
          "curl -fsS http://localhost:$${LAT_WORKERS_HEALTH_PORT:-9090}/health || exit 1",
        ]
      interval: 15s
      timeout: 5s
      retries: 10
      start_period: 40s
    deploy:
      replicas: 1
      restart_policy:
        condition: any
        delay: 5s

  workflows:
    <<: *app
    image: latitudedata/workflows:${LAT_IMAGE_TAG:-latest}
    healthcheck:
      test:
        [
          "CMD-SHELL",
          "curl -fsS http://localhost:$${LAT_WORKFLOWS_HEALTH_PORT:-9091}/health || exit 1",
        ]
      interval: 15s
      timeout: 5s
      retries: 10
      start_period: 40s
    deploy:
      replicas: 1
      restart_policy:
        condition: any
        delay: 5s

  # One-shot: run Postgres (Drizzle Kit) + ClickHouse (goose) migrations, then exit.
  # The application services above wait for this to complete successfully.
  migrations:
    image: latitudedata/migrations:${LAT_IMAGE_TAG:-latest}
    env_file:
      - .env.production
    logging: *logging
    restart: "no"
    depends_on:
      postgres:
        condition: service_healthy
      clickhouse:
        condition: service_healthy
    deploy:
      restart_policy:
        condition: none

  # ==========================================================================
  # Infrastructure — bundled by default. Each block is independently removable;
  # comment one out and point the matching LAT_* env at your existing instance.
  # ==========================================================================

  # --- Postgres (primary OLTP store + pgvector) -----------------------------
  # BYO: remove this block and set LAT_DATABASE_URL / LAT_ADMIN_DATABASE_URL at
  # an existing Postgres (needs the `vector` extension + a schema-create admin).
  postgres:
    image: pgvector/pgvector:pg16
    restart: unless-stopped
    env_file:
      - .env.production
    command: postgres -c wal_level=logical
    volumes:
      - postgres_data:/var/lib/postgresql/data
      - ./docker/init-db.sh:/docker-entrypoint-initdb.d/init-db.sh:ro
    healthcheck:
      test: ["CMD-SHELL", 'pg_isready -U "$$POSTGRES_USER" -d "$$POSTGRES_DB"']
      interval: 10s
      timeout: 5s
      retries: 10
      start_period: 30s
    logging: *logging
    deploy:
      restart_policy:
        condition: any

  # --- ClickHouse (OLAP span/telemetry store) -------------------------------
  # BYO: remove this block and set LAT_CLICKHOUSE_URL / LAT_CLICKHOUSE_MIGRATION_URL etc.
  clickhouse:
    image: clickhouse/clickhouse-server:26.2
    restart: unless-stopped
    env_file:
      - .env.production
    volumes:
      - clickhouse_data:/var/lib/clickhouse
      - ./docker/clickhouse/storage.xml:/etc/clickhouse-server/config.d/storage.xml:ro
    ulimits:
      nofile:
        soft: 262144
        hard: 262144
    healthcheck:
      test:
        [
          "CMD-SHELL",
          'clickhouse-client --user "$$CLICKHOUSE_USER" --password "$$CLICKHOUSE_PASSWORD" --query ''SELECT 1''',
        ]
      interval: 10s
      timeout: 5s
      retries: 10
      start_period: 30s
    logging: *logging
    deploy:
      restart_policy:
        condition: any

  # --- Redis: cache (distributed locks, rate limiters, budgets) -------------
  # BYO: remove this block and set LAT_REDIS_HOST / LAT_REDIS_PORT.
  redis:
    image: redis:7
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 10s
      timeout: 3s
      retries: 10
    logging: *logging
    deploy:
      restart_policy:
        condition: any

  # --- Redis: BullMQ queue (durable; appendonly + noeviction) ---------------
  # BYO: remove this block and set LAT_BULLMQ_HOST / LAT_BULLMQ_PORT.
  redis-bullmq:
    image: redis:7
    restart: unless-stopped
    command: redis-server --maxmemory-policy noeviction --appendonly yes
    volumes:
      - redis_bullmq_data:/data
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 10s
      timeout: 3s
      retries: 10
    logging: *logging
    deploy:
      restart_policy:
        condition: any

  # --- Temporal (workflow engine; Postgres-backed, no Elasticsearch) --------
  # auto-setup creates the `temporal` + `temporal_visibility` databases on the
  # bundled Postgres and the `default` namespace. BYO / Temporal Cloud: remove
  # this block and set LAT_TEMPORAL_ADDRESS / _NAMESPACE / _API_KEY.
  temporal:
    image: temporalio/auto-setup:1.27.2
    restart: unless-stopped
    depends_on:
      postgres:
        condition: service_healthy
    environment:
      DB: postgres12
      DB_PORT: 5432
      POSTGRES_USER: ${POSTGRES_USER:-latitude}
      POSTGRES_PWD: ${POSTGRES_PASSWORD:-secret}
      POSTGRES_SEEDS: postgres
      DBNAME: temporal
      VISIBILITY_DBNAME: temporal_visibility
    logging: *logging
    deploy:
      restart_policy:
        condition: any

  # --- SeaweedFS (Apache-2.0 object store, S3 gateway on :8333) -------------
  # BYO managed S3: remove this block, set LAT_STORAGE_S3_* at your bucket and
  # drop LAT_STORAGE_S3_ENDPOINT / _FORCE_PATH_STYLE. Or use the `fs` driver on
  # a shared volume (set LAT_STORAGE_DRIVER=fs).
  seaweedfs:
    image: chrislusf/seaweedfs:4.32
    restart: unless-stopped
    env_file:
      - .env.production
    entrypoint: ["/bin/sh", "/init.sh"]
    command: ["server", "-dir=/data", "-s3", "-s3.port=8333"]
    volumes:
      - seaweedfs_data:/data
      - ./docker/seaweedfs/init.sh:/init.sh:ro
    logging: *logging
    deploy:
      restart_policy:
        condition: any

volumes:
  postgres_data:
  clickhouse_data:
  redis_bullmq_data:
  seaweedfs_data: