compose.yml

services:
  shepherd_server:
    container_name: shepherd_server
    environment:
      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:-supersecretpassw0rd}
      # The server handles every client request (sync-query status polling,
      # /callback lookups) so it needs a bigger Postgres pool than the
      # workers' default of 10. Budget: each container has its own pool, so
      # keep (sum of pool max sizes) under Postgres max_connections below.
      POSTGRES_POOL_MIN_SIZE: 10
      POSTGRES_POOL_MAX_SIZE: 30
    build:
      context: .
      dockerfile: shepherd_server/Dockerfile
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
    ports:
      - '5439:5439'
  shepherd_db:
    container_name: shepherd_db
    environment:
      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:-supersecretpassw0rd}
    # Connection budget: every container holds its own psycopg pool, so the
    # fleet ceiling is (containers x pool max size). 22 workers/monitor at the
    # default max of 10, plus shepherd_server at 30, is 250 -- which overran
    # the old 200 and surfaced as "sorry, too many clients already" under load
    # even though the comments in this file and shepherd_utils/config.py both
    # say the sum must stay under it. 300 restores the headroom the docs claim.
    # Adding a worker service? Re-do this arithmetic.
    command: postgres -c max_connections=300
    build:
      context: .
      dockerfile: shepherd_db/Dockerfile
    ports:
      - "5432:5432"
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -h 127.0.0.1 -U postgres -d postgres"]
      interval: 2s
      timeout: 2s
      retries: 5
      start_period: 60s
    volumes:
      - ./pgdata:/var/lib/postgresql/data
  shepherd_broker:
    container_name: shepherd_broker
    build:
      context: .
      dockerfile: shepherd_broker/Dockerfile
    environment:
      REDIS_PASSWORD: ${REDIS_PASSWORD:-supersecretpassword}
    ports:
      - "6379:6379"
    restart: unless-stopped
    # The default ``redis-cli ping`` healthcheck silently passed even before
    # Redis was ready, because (a) it didn't authenticate (NOAUTH error still
    # exits 0) and (b) during AOF/RDB load Redis answers ``LOADING`` instead
    # of ``PONG``, which redis-cli also exits 0 on. Grepping for an exact
    # ``PONG`` makes the check actually wait until Redis is fully online.
    # ``start_period`` gives a large RDB time to load before failures count.
    healthcheck:
      test:
        - "CMD-SHELL"
        - 'redis-cli -a "$$REDIS_PASSWORD" --no-auth-warning ping | grep -qx PONG'
      interval: 5s
      timeout: 5s
      retries: 10
      start_period: 60s
    volumes:
      - ./data:/data
  shepherd_broker_insight:
    image: redislabs/redisinsight:latest
    container_name: redis-insight
    restart: unless-stopped
    ports:
      - '5540:5540'
    volumes:
      - ./data/redisinsight:/db
  shepherd_monitor:
    container_name: shepherd_monitor
    build:
      context: .
      dockerfile: workers/monitor/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    ports:
      - "5440:5440"
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
      - ./workers/monitor/monitor_alerts.yaml:/app/monitor_alerts.yaml:ro

  jaeger:
    image: jaegertracing/all-in-one:1.55
    ports:
      - "16686:16686" # UI
      - "4317:4317" # OTLP gRPC
      - "6381:6381/udp" # Agent UDP (Jaeger native)
    environment:
      - COLLECTOR_OTLP_ENABLED=true
      - LOG_LEVEL=error
  

  ########## Shared Workers
  merge_message:
    container_name: merge_message
    build:
      context: .
      dockerfile: workers/merge_message/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  sort_results_score:
    container_name: sort_results_score
    build:
      context: .
      dockerfile: workers/sort_results_score/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  filter_results_top_n:
    container_name: filter_results_top_n
    build:
      context: .
      dockerfile: workers/filter_results_top_n/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  filter_kgraph_orphans:
    container_name: filter_kgraph_orphans
    build:
      context: .
      dockerfile: workers/filter_kgraph_orphans/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  finish_query:
    container_name: finish_query
    build:
      context: .
      dockerfile: workers/finish_query/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  filter_analyses_top_n:
    container_name: filter_analyses_top_n
    build:
      context: .
      dockerfile: workers/filter_analyses_top_n/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  score_paths:
    container_name: score_paths
    build:
      context: .
      dockerfile: workers/score_paths/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
      # First run? Set PATHFINDER_EMBEDDINGS_URL in your .env to a .tar.gz and
      # the worker downloads + extracts the embeddings LMDB into this mount on
      # startup (see README "Worker data (LMDB / sqlite) downloads").
      - ./pathfinder_embeddings:/app/pathfinder_embeddings

  ######### Example ARA
  example_ara:
    container_name: example_ara
    build:
      context: .
      dockerfile: workers/example_ara/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  example_lookup:
    container_name: example_lookup
    build:
      context: .
      dockerfile: workers/example_lookup/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  example_score:
    container_name: example_score
    build:
      context: .
      dockerfile: workers/example_score/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env


  ######### ARAGORN
  aragorn:
    container_name: aragorn
    build:
      context: .
      dockerfile: workers/aragorn/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  aragorn_lookup:
    container_name: aragorn_lookup
    build:
      context: .
      dockerfile: workers/aragorn_lookup/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  aragorn_pathfinder:
    container_name: aragorn_pathfinder
    build:
      context: .
      dockerfile: workers/aragorn_pathfinder/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  aragorn_omnicorp:
    container_name: aragorn_omnicorp
    build:
      context: .
      dockerfile: workers/aragorn_omnicorp/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
      # First run? Set OMNICORP_LMDB_URL in your .env to a .tar.gz and the worker
      # downloads + extracts curies.lmdb / shared_counts.lmdb into this mount on
      # startup (see README "Worker data (LMDB / sqlite) downloads").
      - ./omnicorp_lmdb:/app/omnicorp_lmdb
  aragorn_score:
    container_name: aragorn_score
    build:
      context: .
      dockerfile: workers/aragorn_score/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env


  ######### ARAX
  arax:
    container_name: arax
    build:
      context: .
      dockerfile: workers/arax/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  arax_pathfinder:
    container_name: arax_pathfinder
    build:
      context: .
      dockerfile: workers/arax_pathfinder/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
      # First run? The worker downloads its two sqlite dbs from
      # kg2webhost.rtx.ai, plus the ARAX blocked-concept list from GitHub, on
      # startup -- no credentials needed. All three live here, so they persist
      # across restarts instead of being re-fetched by each new container. Set
      # ARAX_PATHFINDER_TIER_VERSION in your .env if you need a tier other than
      # the default (see README "Worker data (LMDB / sqlite) downloads").
      # Deploying elsewhere? ARAX_PATHFINDER_DBS_DIR must point at the mount --
      # it defaults to a path relative to /app, and a mismatch silently
      # re-downloads everything instead of using the mounted copies.
      - ./arax_pathfinder_dbs:/app/arax_pathfinder_dbs

  arax_rank:
    container_name: arax_rank
    build:
      context: .
      dockerfile: workers/arax_rank/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env

######### BTE
  bte:
    container_name: bte
    build:
      context: .
      dockerfile: workers/bte/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  
  bte_lookup:
    container_name: bte_lookup
    build:
      context: .
      dockerfile: workers/bte_lookup/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env
  
  ######### SIPR
  sipr:
    container_name: sipr
    build:
      context: .
      dockerfile: workers/sipr/Dockerfile
    restart: unless-stopped
    depends_on:
      shepherd_db:
        condition: service_healthy
      shepherd_broker:
        condition: service_healthy
    volumes:
      - ./logs:/app/logs
      - ./.env:/app/.env