Python和RabbitMQ - 从多个渠道收听消费事件的最佳方式?

时间:2015-02-16 20:52:49

标签: python rabbitmq pika

我有两个独立的RabbitMQ实例。我正试图找到听取两者事件的最佳方式。

例如,我可以使用以下内容消费事件:

credentials = pika.PlainCredentials(user, pass)
connection = pika.BlockingConnection(pika.ConnectionParameters(host="host1", credentials=credentials))
channel = connection.channel()
result = channel.queue_declare(Exclusive=True)
self.channel.queue_bind(exchange="my-exchange", result.method.queue, routing_key='*.*.*.*.*')
channel.basic_consume(callback_func, result.method.queue, no_ack=True)
self.channel.start_consuming()

我有第二个主持人,“host2”,我也想听。我想创建两个单独的线程来做到这一点,但从我读过的,pika不是线程安全的。有没有更好的办法?或者会创建两个单独的线程,每个线程都可以监听不同的Rabbit实例(host1和host2)吗?

3 个答案:

答案 0 :(得分:34)

“什么是最好的方式”的答案在很大程度上取决于您的队列使用模式以及“最佳”的含义。由于我还不能对问题发表评论,我只想尝试提出一些可能的解决方案。

在每个例子中,我都假设交换已经宣布。

线程

您可以使用pika在单个进程中使用不同主机上的两个队列使用消息。

你是对的 - 如its own FAQ statespika不是线程安全的,但它可以通过创建每个线程的RabbitMQ主机连接以多线程方式使用。使用threading模块在​​线程中运行此示例如下所示:

import pika
import threading


class ConsumerThread(threading.Thread):
    def __init__(self, host, *args, **kwargs):
        super(ConsumerThread, self).__init__(*args, **kwargs)

        self._host = host

    # Not necessarily a method.
    def callback_func(self, channel, method, properties, body):
        print("{} received '{}'".format(self.name, body))

    def run(self):
        credentials = pika.PlainCredentials("guest", "guest")

        connection = pika.BlockingConnection(
            pika.ConnectionParameters(host=self._host,
                                      credentials=credentials))

        channel = connection.channel()

        result = channel.queue_declare(exclusive=True)

        channel.queue_bind(result.method.queue,
                           exchange="my-exchange",
                           routing_key="*.*.*.*.*")

        channel.basic_consume(self.callback_func,
                              result.method.queue,
                              no_ack=True)

        channel.start_consuming()


if __name__ == "__main__":
    threads = [ConsumerThread("host1"), ConsumerThread("host2")]
    for thread in threads:
        thread.start()

我已经将callback_func声明为纯粹在打印邮件正文时使用ConsumerThread.name的方法。它也可能是ConsumerThread类之外的函数。

过程

或者,您始终可以使用每个要使用事件的队列的使用者代码运行一个进程。

import pika
import sys


def callback_func(channel, method, properties, body):
    print(body)


if __name__ == "__main__":
    credentials = pika.PlainCredentials("guest", "guest")

    connection = pika.BlockingConnection(
        pika.ConnectionParameters(host=sys.argv[1],
                                  credentials=credentials))

    channel = connection.channel()

    result = channel.queue_declare(exclusive=True)

    channel.queue_bind(result.method.queue,
                       exchange="my-exchange",
                       routing_key="*.*.*.*.*")

    channel.basic_consume(callback_func, result.method.queue, no_ack=True)

    channel.start_consuming()

然后运行:

$ python single_consume.py host1
$ python single_consume.py host2  # e.g. on another console

如果您对来自队列的消息所做的工作是CPU-heavy,并且只要CPU中的核心数量> =消费者数量,通常最好使用此方法 - 除非您的队列是大部分时间都是空的,消费者不会利用这个CPU时间*。

异步

另一种选择是涉及一些异步框架(例如Twisted)并在单线程中运行整个事物。

您不能再在异步代码中使用BlockingConnection;幸运的是,pika具有Twisted的适配器:

from pika.adapters.twisted_connection import TwistedProtocolConnection
from pika.connection import ConnectionParameters
from twisted.internet import protocol, reactor, task
from twisted.python import log


class Consumer(object):
    def on_connected(self, connection):
        d = connection.channel()
        d.addCallback(self.got_channel)
        d.addCallback(self.queue_declared)
        d.addCallback(self.queue_bound)
        d.addCallback(self.handle_deliveries)
        d.addErrback(log.err)

    def got_channel(self, channel):
        self.channel = channel

        return self.channel.queue_declare(exclusive=True)

    def queue_declared(self, queue):
        self._queue_name = queue.method.queue

        self.channel.queue_bind(queue=self._queue_name,
                                exchange="my-exchange",
                                routing_key="*.*.*.*.*")

    def queue_bound(self, ignored):
        return self.channel.basic_consume(queue=self._queue_name)

    def handle_deliveries(self, queue_and_consumer_tag):
        queue, consumer_tag = queue_and_consumer_tag
        self.looping_call = task.LoopingCall(self.consume_from_queue, queue)

        return self.looping_call.start(0)

    def consume_from_queue(self, queue):
        d = queue.get()

        return d.addCallback(lambda result: self.handle_payload(*result))

    def handle_payload(self, channel, method, properties, body):
        print(body)


if __name__ == "__main__":
    consumer1 = Consumer()
    consumer2 = Consumer()

    parameters = ConnectionParameters()
    cc = protocol.ClientCreator(reactor,
                                TwistedProtocolConnection,
                                parameters)
    d1 = cc.connectTCP("host1", 5672)
    d1.addCallback(lambda protocol: protocol.ready)
    d1.addCallback(consumer1.on_connected)
    d1.addErrback(log.err)

    d2 = cc.connectTCP("host2", 5672)
    d2.addCallback(lambda protocol: protocol.ready)
    d2.addCallback(consumer2.on_connected)
    d2.addErrback(log.err)

    reactor.run()

这种方法会更好,您消耗的队列越多,消费者执行的工作的CPU限制就越少。

Python 3

由于您已经提到pika,因此我限制自己使用基于Python 2.x的解决方案,因为pika尚未移植。

但是如果你想转移到> = 3.3,一个可能的选择是使用asyncio和AMQP协议之一(你用RabbitMQ说的协议),例如: asynqpaioamqp

* - 请注意,这些是非常浅的提示 - 在大多数情况下,选择并不那么明显;什么对你最好的取决于队列“饱和度”(消息/时间),你收到这些消息后做了什么工作,你的消费者在哪个环境等;除了对所有实现进行基准测试之外,没有办法确定

答案 1 :(得分:1)

以下是我如何使用一个rabbitmq实例同时收听2个队列的示例:

import pika
import threading

threads=[]
def client_info(channel):    
   channel.queue_declare(queue='proxy-python')
   print (' [*] Waiting for client messages. To exit press CTRL+C')


   def callback(ch, method, properties, body):
       print (" Received %s" % (body))

   channel.basic_consume(callback, queue='proxy-python', no_ack=True)
   channel.start_consuming()

def scenario_info(channel):    
   channel.queue_declare(queue='savi-virnet-python')
   print (' [*] Waiting for scenrio messages. To exit press CTRL+C')


   def callback(ch, method, properties, body):
      print (" Received %s" % (body))

   channel.basic_consume(callback, queue='savi-virnet-python', no_ack=True)
   channel.start_consuming()

def manager():
   connection1= pika.BlockingConnection(pika.ConnectionParameters
  (host='localhost'))
   channel1 = connection1.channel()
  connection2= pika.BlockingConnection(pika.ConnectionParameters
  (host='localhost'))
   channel2 = connection2.channel()
   t1 = threading.Thread(target=client_info, args=(channel1,))
   t1.daemon = True
   threads.append(t1)
   t1.start()  

   t2 = threading.Thread(target=scenario_info, args=(channel2,))
   t2.daemon = True
   threads.append(t2)


   t2.start()
   for t in threads:
     t.join()


 manager()

答案 2 :(得分:0)

import asyncio
import tornado.ioloop
import tornado.web

from aio_pika import connect_robust, Message

tornado.ioloop.IOLoop.configure("tornado.platform.asyncio.AsyncIOLoop")
io_loop = tornado.ioloop.IOLoop.current()
asyncio.set_event_loop(io_loop.asyncio_loop)

QUEUE = asyncio.Queue()


class SubscriberHandler(tornado.web.RequestHandler):
    async def get(self):
        message = await QUEUE.get()
        self.finish(message.body)


class PublisherHandler(tornado.web.RequestHandler):
    async def post(self):
        connection = self.application.settings["amqp_connection"]
        channel = await connection.channel()
        try:
            await channel.default_exchange.publish(
                Message(body=self.request.body), routing_key="test",
            )
        finally:
            await channel.close()
            print('ok')
        self.finish("OK")

async def make_app():
    amqp_connection = await connect_robust()
    channel = await amqp_connection.channel()
    queue = await channel.declare_queue("test", auto_delete=True)
    await queue.consume(QUEUE.put, no_ack=True)
    return tornado.web.Application(
        [(r"/publish", PublisherHandler), (r"/subscribe", SubscriberHandler)],
        amqp_connection=amqp_connection,
    )

if __name__ == "__main__":
    app = io_loop.asyncio_loop.run_until_complete(make_app())
    app.listen(8888)
    tornado.ioloop.IOLoop.current().start()

您可以以异步方式使用 aio-pika 更多例子在这里 https://buildmedia.readthedocs.org/media/pdf/aio-pika/latest/aio-pika.pdf

快乐编码:)