Tornado + Motor 微服务架构（Docker + 测试 + Kubernetes）

一、项目结构

复制代码

project-root/
├── docker-compose.yml
├── k8s/                    # Kubernetes 配置文件
│   ├── api-deployment.yaml
│   ├── api-service.yaml
│   ├── user-deployment.yaml
│   ├── user-service.yaml
│   ├── mongo-deployment.yaml
│   ├── mongo-service.yaml
│   └── ingress.yaml
├── common/
│   ├── config.py
│   ├── db.py
│   ├── utils.py
├── api_service/
│   ├── Dockerfile
│   ├── main.py
│   ├── handlers/
│   │   └── user_handler.py
│   ├── services/
│   │   └── user_service.py
│   └── tests/
│       └── test_user.py
├── user_service/
│   ├── Dockerfile
│   ├── main.py
│   ├── handlers/
│   │   └── user_handler.py
│   ├── services/
│   │   └── user_service.py
│   └── tests/
│       └── test_user.py
└── requirements.txt

二、核心依赖（requirements.txt）

复制代码

tornado==6.4
motor==3.3.2
pytest==8.2.0
pytest-asyncio==0.24.0
requests==2.32.3

三、公共模块（common/）

config.py

python 复制代码

import os

MONGO_URI = os.getenv('MONGO_URI', 'mongodb://mongo:27017')
DB_NAME = os.getenv('DB_NAME', 'microdb')
SERVICE_NAME = os.getenv('SERVICE_NAME', 'default-service')
PORT = int(os.getenv('PORT', 8000))

db.py

python 复制代码

from motor.motor_tornado import MotorClient
from common.config import MONGO_URI, DB_NAME

client = MotorClient(MONGO_URI)
db = client[DB_NAME]

utils.py

python 复制代码

import json
import tornado.web

class BaseHandler(tornado.web.RequestHandler):
    def set_default_headers(self):
        self.set_header('Content-Type', 'application/json')

    def write_json(self, data, status=200):
        self.set_status(status)
        self.write(json.dumps(data))

四、API 服务（api_service/）

Dockerfile

Dockerfile 复制代码

FROM python:3.11-slim
WORKDIR /app
COPY ../../requirements.txt ./
RUN pip install -r requirements.txt
COPY . .
CMD ["python", "main.py"]

main.py

python 复制代码

import tornado.ioloop
import tornado.web
from tornado.process import fork_processes
from handlers.user_handler import UserHandler
from common.config import PORT


def make_app():
    return tornado.web.Application([
        (r"/user", UserHandler),
    ])

if __name__ == '__main__':
    fork_processes(None)  # 多进程 worker
    app = make_app()
    app.listen(PORT)
    tornado.ioloop.IOLoop.current().start()

handlers/user_handler.py

python 复制代码

from common.utils import BaseHandler
from services.user_service import UserService

class UserHandler(BaseHandler):
    async def get(self):
        users = await UserService().get_all_users()
        self.write_json({"users": users})

    async def post(self):
        data = self.request.body.decode()
        result = await UserService().create_user(data)
        self.write_json(result)

services/user_service.py

python 复制代码

from common.db import db
import json

class UserService:
    async def get_all_users(self):
        cursor = db.users.find()
        return [doc async for doc in cursor]

    async def create_user(self, data):
        user = json.loads(data)
        result = await db.users.insert_one(user)
        return {"_id": str(result.inserted_id)}

tests/test_user.py

python 复制代码

import pytest
import asyncio
from services.user_service import UserService

@pytest.mark.asyncio
async def test_create_user():
    service = UserService()
    res = await service.create_user('{"name":"test_user"}')
    assert "_id" in res

五、User 服务（user_service/）

结构相同，只是接口路径不同，例如 /profile。

六、Docker Compose

docker-compose.yml

yaml 复制代码

version: '3.8'
services:
  mongo:
    image: mongo:6.0
    container_name: mongo
    ports:
      - "27017:27017"

  api_service:
    build: ./api_service
    container_name: api_service
    environment:
      - MONGO_URI=mongodb://mongo:27017
      - DB_NAME=microdb
      - PORT=8000
    ports:
      - "8000:8000"
    depends_on:
      - mongo

  user_service:
    build: ./user_service
    container_name: user_service
    environment:
      - MONGO_URI=mongodb://mongo:27017
      - DB_NAME=microdb
      - PORT=8001
    ports:
      - "8001:8001"
    depends_on:
      - mongo

七、Kubernetes 配置示例（k8s/）

api-deployment.yaml

yaml 复制代码

apiVersion: apps/v1
kind: Deployment
metadata:
  name: api-deployment
spec:
  replicas: 2
  selector:
    matchLabels:
      app: api-service
  template:
    metadata:
      labels:
        app: api-service
    spec:
      containers:
        - name: api-service
          image: api_service:latest
          ports:
            - containerPort: 8000

api-service.yaml

yaml 复制代码

apiVersion: v1
kind: Service
metadata:
  name: api-service
spec:
  type: ClusterIP
  selector:
    app: api-service
  ports:
    - port: 8000
      targetPort: 8000

ingress.yaml

yaml 复制代码

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: micro-ingress
spec:
  rules:
    - host: micro.local
      http:
        paths:
          - path: /user
            pathType: Prefix
            backend:
              service:
                name: api-service
                port:
                  number: 8000

八、运行与测试

启动开发环境：
bash 复制代码
```
docker-compose up --build
```
本地运行测试：
bash 复制代码
```
pytest -v
```
部署到 Kubernetes：
bash 复制代码
```
kubectl apply -f k8s/
```

九、说明

Tornado 使用多进程 worker（fork_processes(None)）自动创建与 CPU 核数相同的进程。
MongoDB 异步连接由 Motor 驱动支持。
测试通过 pytest-asyncio 实现异步单元测试。
Kubernetes YAML 已含 Ingress，可直接对外暴露 API 服务。

这一结构可直接扩展为完整的微服务体系，每个服务都独立运行、可水平扩展。