python - 如何让 Rabbitmq 队列以存储转发模式运行？

Question

我正在使用 Python Pika 客户端试验 Rabbit MQ。我想做的是让我的 AMQP 发送器在存储和转发模式下运行，即如果服务器或网络出现故障，能够开始排队消息，并在以后可靠地传递它们。我怎么做？我的 amqp-sender.py 代码如下:

    import pika
    import psutil
    import time
    import datetime
    import log
    import json
    import logging
    import uuid
    from dateutil.tz import tzlocal

    logging.basicConfig()
    logger = log.setup_custom_logger('amqp_send', 'amqp_send.log')

    connection = pika.BlockingConnection(pika.ConnectionParameters(host='54.191.161.213'))
    channel = connection.channel()
    channel.confirm_delivery()

    channel.queue_declare(queue='ems.data')

    def get_mac_address():
        return ':'.join(['{:02x}'.format((uuid.getnode() >> i) & 0xff) for i in range(0,8*6,8)][::-1])

    while True:
        now = datetime.datetime.now(tzlocal())
        timestamp = now.strftime('%Y-%m-%d %H:%M:%S.%f %z')
        data = {
            'timestamp':timestamp,
            'systemId':get_mac_address(),
            'cpuPct':psutil.cpu_percent(),
            'memoryUsed':psutil.virtual_memory().used
        }
        msg=json.dumps(data)
        delivered=channel.basic_publish(exchange='', routing_key='abc', body=msg, mandatory=True)
        if delivered:
            logger.info("delivered %s" % msg)
        else:
            logger.error('failed to deliver %s' % msg)
        time.sleep(1)

    connection.close()





Answer 1

您需要将channel.confirm_delivery()与mandatory=True一起使用

confirm_delivery 将返回 bool 值，具体取决于 Rabbit 是否正确处理消息。

强制标志:

This flag tells the server how to react if a message cannot be routed to a queue. Specifically, if mandatory is set and after running the bindings the message was placed on zero queues then the message is returned to the sender (with a basic.return). If mandatory had not been set under the same circumstances the server would silently drop the message.

所以你会得到这样的东西:

channel.confirm_delivery()
delivered = channel.basic_publish(exchange='', routing_key='ems.data', body=msg, mandatory=True)
if not delivered:
    # store message for later reprocessing