Implementation:Infiniflow Ragflow Entrypoint And Server

Knowledge Sources	RAGFlow
Domains	DevOps, Backend_Architecture
Last Updated	2026-02-12 06:00 GMT

Overview

Concrete tool for starting the RAGFlow server and workers provided by the entrypoint script and ragflow_server.py.

Description

docker/entrypoint.sh (273 lines) orchestrates multi-process startup: envsubst, nginx, ragflow_server.py, sync_data_source.py, admin_server.py (optional), task_executor.py workers (with jemalloc), and MCP server (optional). api/ragflow_server.py initializes Flask, calls init_web_db and init_web_data for database schema and seed data, then starts the Quart/Hypercorn server.

Usage

Automatic in Docker. For local dev: bash docker/launch_backend_service.sh.

Code Reference

Source Location

Repository: ragflow
File: docker/entrypoint.sh (L1-273), api/ragflow_server.py (L76-157)

Signature

# api/ragflow_server.py
def main():
    """Start the RAGFlow server.

    CLI args:
        --enable-adminserver
        --disable-webserver
        --disable-taskexecutor
        --disable-datasync
        --enable-mcpserver
        --workers=N (default 1)
        --host-id=STRING

    Server config:
        HOST_IP: default "0.0.0.0"
        HOST_PORT: default 9380
    """

Import

# Entry point script - not importable
# Run: python3 api/ragflow_server.py

I/O Contract

Inputs

Name	Type	Required	Description
service_conf.yaml	file	Yes	Generated configuration
MySQL connection	service	Yes	Database must be healthy

Outputs

Name	Type	Description
HTTP server	service	Port 9380 (API), 80 (HTTP via nginx)
Task workers	processes	Background document processors
Admin server	service	Optional, port 9381
MCP server	service	Optional, port 9382

Usage Examples

# Docker startup (automatic)
docker compose up -d
docker logs -f ragflow-server

# Local development
source .venv/bin/activate
export PYTHONPATH=$(pwd)
bash docker/launch_backend_service.sh

Related Pages

Implements Principle

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment