如何在 Python 中从同一个脚本运行 2 个服务器

Question

我正在构建一个解决方案，该解决方案依赖于来自现场 IOT 设备的发布/订阅事件来启动一些处理。出于我的目的，我使用 Flask 来运行我的主应用程序。然后我运行 mqtt 客户端连接到 MQTT 服务器并监听事件。

当我使用python app.py 运行我的主 Flask app.py 时，它会同时启动 Flask 服务和 MQTT 客户端。但是，当我尝试使用 gunicorn 运行时，它只启动了 Flask 服务，而没有启动 MQTT 客户端。

我可以使用什么生产级服务（如果不是 gunicorn）来运行这两个服务以及如何使用它？

import logging, json, requests, os
import paho.mqtt.client as mqtt

from flask import Flask, request, url_for
from flask_restful import reqparse, abort, Api, Resource
from logging.handlers import SysLogHandler
from logging import StreamHandler
from flask.ext.superadmin import Admin, BaseView, model
from redis import Redis
from rq import Queue
from task import Task
from dateutil.relativedelta import relativedelta
from datetime import datetime, date
..
..
from commons import db, bootstrap

app = Flask(__name__)
api = Api(app)

bootstrap.bootstrap_app(app)

#Setup (Redis) Queue store
q = Queue(connection=Redis())

#Intialize task instance
task = Task(app)
..
...
@app.before_request
def log_request_info():
    app.logger.debug('Headers: %s', request.headers)
    app.logger.debug('Body: %s', request.get_data())

#API resource routing
api.add_resource(Test, '/test')


def session_report(client, userdata, message):
    print message.topic, message.payload

    # Redirect to session_report endpoint
    with app.test_request_context():
        url = 'http://localhost:5000'+url_for('sessionreport', _external=False)
        headers = {'content-type': 'application/json'}
        response = requests.request("POST", url, data=message.payload, 
            headers=headers)

def process_ack(client, userdata, message):
    try:
        user = User()
        user.update_account(account_number=next(iter(json.loads(message.payload))))
    except Exception, e:
        print e

def publish_accounts(client):
    user_obj = User()
    users = {user.id: user.account_number for user in user_obj.get_users()}
    client.publish('accounts', payload=json.dumps(users), qos=1, retain=True)


# paho callbacks

def on_connect(client, userdata, flags, rc):
    print "CONNECTED!", str(rc)

    # Subscribe to topis(s) here
    client.subscribe("mine/#")
    client.subscribe("session/#")
    client.subscribe("ack")

    # Add callbacks to subscribed topics
    client.message_callback_add("session/#", session_report)
    client.message_callback_add("ack", process_ack)

    # Publish latest list of accounts
    publish_accounts(client)

def on_subscribe(client, userdata, mid, granted_qos):
    print "Subscribed: ", str(mid), str(granted_qos)

def on_message(client, userdata, msg):
    print msg.topic, msg.payload

def on_publish(client, userdata, mid):
    print "PUBLISHED!"
    app.logger.info('PUBLISHED : {} -- {}'.format(mid, userdata))


if __name__ == '__main__':
    handler = StreamHandler()   #SysLogHandler()
    handler.setLevel(logging.DEBUG)
    app.logger.addHandler(handler)  
    client = mqtt.Client()
    client.on_connect = on_connect
    client.on_subscribe = on_subscribe
    client.on_message = on_message
    client.connect("localhost", 1883, 60)
    client.loop_start()
    app.run(debug=True, host='0.0.0.0', port=5000)

Answer 1

Flask 只是 Python 的一个微框架。 如果要通过python同时运行多个方法，则需要使用多线程编程。

我又看了一些 Flask 文档，发现 Flask 支持多线程并发。

试试看：

app.run(host='0.0.0.0',port=5000,debug=True, threaded = True)