docker-compose.dev.yml

# SynapseOps — Docker Compose Dev
# Solo infraestructura. Backend, ml-engine y frontend corren en IDEs.
# Uso: docker compose -f docker-compose.dev.yml up -d

name: synapseops

services:

  postgres-db:
    image: postgres:17-alpine
    container_name: postgres-db
    restart: unless-stopped
    ports:
      - "5433:5432"           
    environment:
      POSTGRES_DB: orchestrator
      POSTGRES_USER: ${DB_USERNAME}
      POSTGRES_PASSWORD: ${DB_PASSWORD}
    volumes:
      - postgres_data:/var/lib/postgresql/data
    networks:
      - mlops-network
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U ${DB_USERNAME} -d orchestrator"]
      interval: 10s
      timeout: 5s
      retries: 5
      start_period: 15s

  kafka-broker:
    image: apache/kafka:3.7.0
    container_name: kafka-broker
    restart: unless-stopped
    ports:
      - "9092:9092"     
      - "9094:9094"     
    environment:
      KAFKA_NODE_ID: 1
      KAFKA_PROCESS_ROLES: broker,controller
      KAFKA_LISTENERS: INTERNAL://0.0.0.0:9092,EXTERNAL://0.0.0.0:9094,CONTROLLER://0.0.0.0:9093
      KAFKA_ADVERTISED_LISTENERS: INTERNAL://kafka-broker:9092,EXTERNAL://localhost:9094
      KAFKA_LISTENER_SECURITY_PROTOCOL_MAP: INTERNAL:PLAINTEXT,EXTERNAL:PLAINTEXT,CONTROLLER:PLAINTEXT
      KAFKA_INTER_BROKER_LISTENER_NAME: INTERNAL
      KAFKA_CONTROLLER_QUORUM_VOTERS: 1@kafka-broker:9093
      KAFKA_CONTROLLER_LISTENER_NAMES: CONTROLLER
      KAFKA_OFFSETS_TOPIC_REPLICATION_FACTOR: 1
      KAFKA_TRANSACTION_STATE_LOG_REPLICATION_FACTOR: 1
      KAFKA_TRANSACTION_STATE_LOG_MIN_ISR: 1
      KAFKA_AUTO_CREATE_TOPICS_ENABLE: "true"
      KAFKA_LOG_RETENTION_HOURS: 24
      CLUSTER_ID: ${KAFKA_CLUSTER_ID}
    volumes:
      - kafka_data:/var/lib/kafka/data
    networks:
      - mlops-network
    healthcheck:
      test: ["CMD-SHELL", "/opt/kafka/bin/kafka-topics.sh --bootstrap-server localhost:9092 --list"]
      interval: 15s
      timeout: 10s
      retries: 5
      start_period: 30s

  mlflow-server:
    image: ghcr.io/mlflow/mlflow:v2.21.3
    container_name: mlflow-server
    restart: unless-stopped
    ports:
      - "5000:5000"
    command: >
      mlflow server
      --backend-store-uri sqlite:////mlflow/mlflow.db
      --default-artifact-root /mlflow/artifacts
      --host 0.0.0.0
      --port 5000
    volumes:
      - mlflow_db:/mlflow
      - mlflow_artifacts:/mlflow/artifacts
    networks:
      - mlops-network
    healthcheck:
      test: ["CMD", "python", "-c",
             "import urllib.request; urllib.request.urlopen('http://localhost:5000/health')"]
      interval: 10s
      timeout: 5s
      retries: 8
      start_period: 60s

volumes:
  postgres_data:
    name: synapseops_postgres_data
  kafka_data:
    name: synapseops_kafka_data
  mlflow_db:
    name: synapseops_mlflow_db
  mlflow_artifacts:
    name: synapseops_mlflow_artifacts
  storage_data:             
    name: synapseops_storage_data  

networks:
  mlops-network:
    name: mlops-network
    driver: bridge