Docker Compose Configuration

In the development environment, the docker-compose.yml file is critical for setting up the various services required for the application. Below is an explanation of each service block along with configuration options:

Redis Service

    redis:
      image: redis:6.0.16
      restart: always
      volumes:
        - ./redis-data:/data
      command: ["redis-server", "--loglevel", "warning"]

• image: Specifies which version of Redis to use.
• restart: Ensures the service restarts automatically on failure.
• volumes: Maps a local directory to the Redis data directory for persistent storage.
• command: Sets the log level of the Redis server to minimize logs during operation.

Backend Service

    backend:
      image: tidbai/backend:0.2.8
      restart: always
      depends_on:
        - redis
      ports:
        - "8000:80"
      env_file:
        - .env
      volumes:
        - ./data:/shared/data
      logging:
        driver: json-file
        options:
          max-size: "50m"
          max-file: "6"

• image: Defines the version of the backend service.
• depends_on: Ensures Redis is started before the backend service.
• ports: Maps port 8000 on the host to port 80 on the container.
• env_file: Indicates that environment variables will be loaded from the .env file.
• volumes: Similar to Redis, it allows data to be shared between the host and the container.
• logging: Configures logging behavior to manage log file sizes.

Frontend Service

    frontend:
      image: tidbai/frontend:0.2.8
      restart: always
      depends_on:
        - backend
      ports:
        - 3000:3000
      environment:
        BASE_URL: http://backend
      logging:
        driver: json-file
        options:
          max-size: "50m"
          max-file: "6"

• image: Specifies the version of the frontend service.
• depends_on: Ensures the backend service is up and running first.
• ports: Exposes port 3000 for the frontend interface.
• environment: Sets the base URL for API requests made from the frontend.

Background Service

    background:
      image: tidbai/backend:0.2.8
      restart: always
      depends_on:
        - redis
      ports:
        - "5555:5555"
      env_file:
        - .env
      volumes:
        - ./data:/shared/data
      command: /usr/bin/supervisord
      logging:
        driver: json-file
        options:
          max-size: "50m"
          max-file: "6"

• command: Overrides the default command to run the Supervisor process, allowing the management of multiple processes.

Local Embedding Reranker Service

    local-embedding-reranker:
      image: tidbai/local-embedding-reranker:v3-with-cache
      ports:
        - 5001:5001
      environment:
        - PRE_LOAD_DEFAULT_EMBEDDING_MODEL=true
        - PRE_LOAD_DEFAULT_RERANKER_MODEL=false
        - TRANSFORMERS_OFFLINE=1

• environment: Activates features related to pre-loading embedding and reranker models, controlling behavior based on whether GPU support is enabled.

Profiles Block

profiles:
  - local-embedding-reranker

• profiles: Allows for specific configurations to be activated, in this case, the local embedding reranker.

Environment Variables

The .env file contains sensitive information and configurations. Below is an example configuration and essential notes:

ENVIRONMENT=production

SECRET_KEY=

TIDB_HOST=xxxxx.prod.aws.tidbcloud.com
TIDB_USER=
TIDB_PASSWORD=
TIDB_DATABASE=
TIDB_SSL=true

EMBEDDING_DIMS=1536
EMBEDDING_MAX_TOKENS=8191

• SECRET_KEY: Generate a secure key for the application using the provided Python command to ensure it meets security requirements.

$ python3 -c "import secrets; print(secrets.token_urlsafe(32))"

• TIDB settings: Fill in your TiDB connection details. TiDB Serverless is highly recommended for simplicity.

• EMBEDDING_DIMS and EMBEDDING_MAX_TOKENS: These configurations are crucial for embedding model specifications. Ensure they are compatible with the embedding model being used.

The configuration steps outlined above should allow for the successful set-up of your development environment using pingcap/autoflow.

Source: Given configuration files and environment variable examples.