一、RabbitMQ消息队列介绍
RabbitMQ是在两个独立得python程序,或其他语言交互时使用。
RabbitMQ:erlang语言 开发的。
Python中连接RabbitMQ的模块:pika 、Celery(分布式任务队列) 、haigha
可以维护很多得队列
RabbitMQ 教程官网:http://www.rabbitmq.com/getstarted.html
几个概念说明:
Broker:简单来说就是消息队列服务器实体。
Exchange:消息交换机,他制定消息按什么规则,路由到哪个队列。
Queue:消息队列载体,每个消息都会被投入一个或多个队列。
Binding:绑定,他的作用就是把exchange和queue按照路由规则绑定起来。
Routing Key:路由关键字,exchange根据这个关键字进行消息投递。
vhost:虚拟主机,一个broker里可以设多个vhost,用作不同用户得权限分离。
producer:消息生产者,就是投递消息得程序。
consumer:消息消费者,就是接受消息得程序。
channel:消息通道,在客户端得每个连接里。可以建立多个channel,每个channel代表一个会话任务。
二、RabbitMQ基本示范
1.Rabbitmq安装
ubuntu系统
install rabbitmq-server # 直接搞定
---
centos系统
1)Install Erlang
1)Install Erlang # For EL5: rpm -Uvh http://download.fedoraproject.org/pub/epel/5/i386/epel-release-5-4.noarch.rpm # For EL6: rpm -Uvh http://download.fedoraproject.org/pub/epel/6/i386/epel-release-6-8.noarch.rpm # For EL7: rpm -Uvh http://download.fedoraproject.org/pub/epel/7/x86_64/e/epel-release-7-8.noarch.rpm yum install erlang
2)Install RabbitMQ Server
rpm --import https://www.rabbitmq.com/rabbitmq-release-signing-key.asc
yum install rabbitmq-server-3.6.5-1.noarch.rpm
3)use RabbitMQ Server
chkconfig rabbitmq-server on
service rabbitmq-server stop/start
2.基本示例子
发送端producer
import pika # 建立一个实例 connection = pika.BlockingConnection( pika.ConnectionParameters('localhost',5672) # 默认端口5672,可不写 ) # 声明一个管道,在管道里发消息 channel = connection.channel() # 在管道里声明queue channel.queue_declare(queue='hello') # RabbitMQ a message can never be sent directly to the queue, it always needs to go through an exchange. channel.basic_publish(exchange='', routing_key='hello', # queue名字 body='Hello World!') # 消息内容 print(" [x] Sent 'Hello World!'") connection.close() # 队列关闭
接收端consumer
import pika import time # 建立实例 connection = pika.BlockingConnection(pick.ConnectionParameters('hocalhost')) # 声明管道 channel = connection.channel() # 为什么声明了一个‘hello’队列? # 如果确定已经声明了,可以不声明。但是你不知道那个机器先运行,所以要声明两次。 # 通常是先运行消费者 channel.queue_declare(queue='hello') def callback(ch, method, properties, body):#四个参数为标准格式 print("[x]Received %r"%body) time.sleep(15) ch.basic_ack(delivry_tay = method.delivery_tay)# 告诉生成者,消息处理完成 channel.basic_consume(# 消费消息 callback, # 如果收到消息,就调用callback函数来处理消息 queue='hello',# 要消费的队列 # no ack=True # 消息确认 # 一般不写。宕机则生产者检测到发给其他消费者 ) print('[*]Waiting for messages.To exit press CTRL+C') channel.start_consuming() # 开始消费消息
3.RabbitMQ消息分发轮询
一个生产者多个消费者
采用轮询机制;把消息依次分发
假如消费者处理雄安熙需要15秒,如果宕机了,那这个消息处理还没有处理完,怎么处理?
(可以模拟消费端断了,分别注释和不注释no_ack=True看一下)
没有回复,就代表消息没有处理完,
上面的效果消费端断了就转到另外一个消费端去了,但是生产者怎么知道消费端断了呢?
因为生产者和消费者是通过socket连接的,socket断了,就说明消费端断开了。
上面的模式只是依次分发,实际情况是机器配置不一样。怎么设置类似权重的操作?
RabbitMQ怎么办呢,RabbitMQ做了简单的处理就能实现公平的分发。
就是RabbitMQ给消费者发消息的时候检测下消费者里的消息数量,如果超过指定值(比如1条),就不给你发了。
只需要在消费者端,channel.basic_consume前加上就可以了。
channel.basic_qos(prefetch_count=1)# 类似权重,按能力分发,如果有一个消息,就不在给你发
channel.basic_consume( # 消费消息
三、Rabbit MQ消息持久化(durable、properties)
1.RabbitMQ相关命令
rabbitmqctl list_queues # 查看当前queue数量及queue里消息数量
2.消息持久化
如果队列里还有消息,RabbitMQ服务端宕机了呢?消息还在不在?
把RabbitMQ服务重启,看一下消息在不在。
上面的情况下,宕机了,消息就久了,下面看看如何把消息持久化。
每次声明队列的时候,都加上durable,注意每个队列都得写,客户端、服务端声明的时候都得写。
在管道里声明queue
channel.queue_declare(queue='hello2', durable=True)
测试结果发现,只是把队列持久化了,但队列里的消息没了。
durable的作用只是把队列持久化。离消息持久话还差一步:
发送端发送消息时,加上properties
import pika import sys connection = pika.BlockingConnection(pika.ConnectionParameters( host='localhost')) channel = connection.channel() channel.exchange_declare(exchange='direct_logs', type='direct') # 重要程度级别,这里默认定义为 info severity = sys.argv[1] if len(sys.argv) > 1 else 'info' message = ' '.join(sys.argv[2:]) or 'Hello World!' channel.basic_publish(exchange='direct_logs', routing_key=severity, body=message) print(" [x] Sent %r:%r" % (severity, message)) connection.close()
接收端subscriber
import pika import sys connection = pika.BlockingConnection(pika.ConnectionParameters( host='localhost')) channel = connection.channel() channel.exchange_declare(exchange='topic_logs', type='topic') routing_key = sys.argv[1] if len(sys.argv) > 1 else 'anonymous.info' message = ' '.join(sys.argv[2:]) or 'Hello World!' channel.basic_publish(exchange='topic_logs', routing_key=routing_key, body=message) print(" [x] Sent %r:%r" % (routing_key, message)) connection.close()
接收端 consumer
import pika import time connection = pika.BlockingConnection(pika.ConnectionParameters( 'localhost')) channel = connection.channel() channel.queue_declare(queue='hello2', durable=True) def callback(ch, method, properties, body): print(" [x] Received %r" % body) time.sleep(10) ch.basic_ack(delivery_tag = method.delivery_tag) # 告诉生产者,消息处理完成 channel.basic_qos(prefetch_count=1) # 类似权重,按能力分发,如果有一个消息,就不在给你发 channel.basic_consume( # 消费消息 callback, # 如果收到消息,就调用callback queue='hello2', # no_ack=True # 一般不写,处理完接收处理结果。宕机则发给其他消费者 ) print(' [*] Waiting for messages. To exit press CTRL+C') channel.start_consuming()
四、RabbitMQ广播模式(exchange)
前面的效果都是一对一发,如果做一个广播效果可不可以,这时候就要用到exchange了
exchange必须精确的知道收到的消息要发给谁。exchange的类型决定了怎么处理,
类型有以下几种:
fanout: 所有绑定到此exchange的queue都可以接收消息
direct: 通过routingKey和exchange决定的那个唯一的queue可以接收消息
topic: 所有符合routingKey(此时可以是一个表达式)的routingKey所bind的queue可以接收消息
1.fanout 纯广播、all
需要queue和exchange绑定,因为消费者不是和exchange直连的,消费者是连在queue上,queue绑定在exchange上,消费者只会在queu里度消息
|------------------------| | /—— queue <—|—> consumer1 producer —|—exchange1 <bind | \ | \—— queue <—|—> consumer2 \-|-exchange2 …… | |------------------------|
发送端 publisher 发布、广播
import pika import sys connection = pika.BlockingConnection(pika.ConnectionParameters( host='localhost')) channel = connection.channel() # 注意:这里是广播,不需要声明queue channel.exchange_declare(exchange='logs', # 声明广播管道 type='fanout') # message = ' '.join(sys.argv[1:]) or "info: Hello World!" message = "info: Hello World!" channel.basic_publish(exchange='logs', routing_key='', # 注意此处空,必须有 body=message) print(" [x] Sent %r" % message) connection.close()
接收端 subscriber 订阅
import pika connection = pika.BlockingConnection(pika.ConnectionParameters( host='localhost')) channel = connection.channel() channel.exchange_declare(exchange='logs', type='fanout') # 不指定queue名字,rabbit会随机分配一个名字,exclusive=True会在使用此queue的消费者断开后,自动将queue删除 result = channel.queue_declare(exclusive=True) # 获取随机的queue名字 queue_name = result.method.queue print("random queuename:", queue_name) channel.queue_bind(exchange='logs', # queue绑定到转发器上 queue=queue_name) print(' [*] Waiting for logs. To exit press CTRL+C') def callback(ch, method, properties, body): print(" [x] %r" % body) channel.basic_consume(callback, queue=queue_name, no_ack=True) channel.start_consuming()
注意:广播,是实时的,收不到就没了,消息不会存下来,类似收音机。
2.direct 有选择的接收消息
接收者可以过滤消息,只收我想要的消息
发送端publisher
import pika import sys connection = pika.BlockingConnection(pika.ConnectionParameters( host='localhost')) channel = connection.channel() channel.exchange_declare(exchange='direct_logs', type='direct') # 重要程度级别,这里默认定义为 info severity = sys.argv[1] if len(sys.argv) > 1 else 'info' message = ' '.join(sys.argv[2:]) or 'Hello World!' channel.basic_publish(exchange='direct_logs', routing_key=severity, body=message) print(" [x] Sent %r:%r" % (severity, message)) connection.close()
接收端subscriber
import pika import sys connection = pika.BlockingConnection(pika.ConnectionParameters( host='localhost')) channel = connection.channel() channel.exchange_declare(exchange='direct_logs', type='direct') result = channel.queue_declare(exclusive=True) queue_name = result.method.queue # 获取运行脚本所有的参数 severities = sys.argv[1:] if not severities: sys.stderr.write("Usage: %s [info] [warning] [error]\n" % sys.argv[0]) sys.exit(1) # 循环列表去绑定 for severity in severities: channel.queue_bind(exchange='direct_logs', queue=queue_name, routing_key=severity) print(' [*] Waiting for logs. To exit press CTRL+C') def callback(ch, method, properties, body): print(" [x] %r:%r" % (method.routing_key, body)) channel.basic_consume(callback, queue=queue_name, no_ack=True) channel.start_consuming()
运行接收端,指定接收级别的参数,例:
python direct_sonsumer.py info warning
python direct_sonsumer.py warning error
3.topic 更细致的过滤
比如把error中,apache和mysql的分别或取出来
发送端publisher
import pika import sys connection = pika.BlockingConnection(pika.ConnectionParameters( host='localhost')) channel = connection.channel() channel.exchange_declare(exchange='topic_logs', type='topic') routing_key = sys.argv[1] if len(sys.argv) > 1 else 'anonymous.info' message = ' '.join(sys.argv[2:]) or 'Hello World!' channel.basic_publish(exchange='topic_logs', routing_key=routing_key, body=message) print(" [x] Sent %r:%r" % (routing_key, message)) connection.close()
接收端 subscriber
import pika import sys connection = pika.BlockingConnection(pika.ConnectionParameters( host='localhost')) channel = connection.channel() channel.exchange_declare(exchange='direct_logs', type='direct') result = channel.queue_declare(exclusive=True) queue_name = result.method.queue # 获取运行脚本所有的参数 severities = sys.argv[1:] if not severities: sys.stderr.write("Usage: %s [info] [warning] [error]\n" % sys.argv[0]) sys.exit(1) # 循环列表去绑定 for severity in severities: channel.queue_bind(exchange='direct_logs', queue=queue_name, routing_key=severity) print(' [*] Waiting for logs. To exit press CTRL+C') def callback(ch, method, properties, body): print(" [x] %r:%r" % (method.routing_key, body)) channel.basic_consume(callback, queue=queue_name, no_ack=True) channel.start_consuming()
运行接收端,指定接收哪些消息,例:
python topic_sonsumer.py *.info python topic_sonsumer.py *.error mysql.* python topic_sonsumer.py '#' # 接收所有消息 # 接收所有的 logs run: # python receive_logs_topic.py "#" # To receive all logs from the facility "kern": # python receive_logs_topic.py "kern.*" # Or if you want to hear only about "critical" logs: # python receive_logs_topic.py "*.critical" # You can create multiple bindings: # python receive_logs_topic.py "kern.*" "*.critical" # And to emit a log with a routing key "kern.critical" type: # python emit_log_topic.py "kern.critical" "A critical kernel error"
4.RabbitMQ RPC 实现(Remote procedure call)
不知道你有没有发现,上面的流都是单向的,如果远程的机器执行完返回结果,就实现不了了。
如果返回,这种模式叫什么呢,RPC(远程过程调用),snmp就是典型的RPC
RabbitMQ能不能返回呢,怎么返回呢?既是发送端又是接收端。
但是接收端返回消息怎么返回?可以发送到发过来的queue里么?不可以。
返回时,再建立一个queue,把结果发送新的queue里
为了服务端返回的queue不写死,在客户端给服务端发指令的的时候,同时带一条消息说,你结果返回给哪个queue
RPC client
import pika import uuid import time class FibonacciRpcClient(object): def __init__(self): self.connection = pika.BlockingConnection(pika.ConnectionParameters( host='localhost')) self.channel = self.connection.channel() result = self.channel.queue_declare(exclusive=True) self.callback_queue = result.method.queue self.channel.basic_consume(self.on_response, # 只要一收到消息就调用on_response no_ack=True, queue=self.callback_queue) # 收这个queue的消息 def on_response(self, ch, method, props, body): # 必须四个参数 # 如果收到的ID和本机生成的相同,则返回的结果就是我想要的指令返回的结果 if self.corr_id == props.correlation_id: self.response = body def call(self, n): self.response = None # 初始self.response为None self.corr_id = str(uuid.uuid4()) # 随机唯一字符串 self.channel.basic_publish( exchange='', routing_key='rpc_queue', # 发消息到rpc_queue properties=pika.BasicProperties( # 消息持久化 reply_to = self.callback_queue, # 让服务端命令结果返回到callback_queue correlation_id = self.corr_id, # 把随机uuid同时发给服务器 ), body=str(n) ) while self.response is None: # 当没有数据,就一直循环 # 启动后,on_response函数接到消息,self.response 值就不为空了 self.connection.process_data_events() # 非阻塞版的start_consuming() # print("no msg……") # time.sleep(0.5) # 收到消息就调用on_response return int(self.response) if __name__ == '__main__': fibonacci_rpc = FibonacciRpcClient() print(" [x] Requesting fib(7)") response = fibonacci_rpc.call(7) print(" [.] Got %r" % response)
RPC server
import pika import time def fib(n): if n == 0: return 0 elif n == 1: return 1 else: return fib(n-1) + fib(n-2) def on_request(ch, method, props, body): n = int(body) print(" [.] fib(%s)" % n) response = fib(n) ch.basic_publish( exchange='', # 把执行结果发回给客户端 routing_key=props.reply_to, # 客户端要求返回想用的queue # 返回客户端发过来的correction_id 为了让客户端验证消息一致性 properties=pika.BasicProperties(correlation_id = props.correlation_id), body=str(response) ) ch.basic_ack(delivery_tag = method.delivery_tag) # 任务完成,告诉客户端 if __name__ == '__main__': connection = pika.BlockingConnection(pika.ConnectionParameters( host='localhost')) channel = connection.channel() channel.queue_declare(queue='rpc_queue') # 声明一个rpc_queue , channel.basic_qos(prefetch_count=1) # 在rpc_queue里收消息,收到消息就调用on_request channel.basic_consume(on_request, queue='rpc_queue') print(" [x] Awaiting RPC requests") channel.start_consuming()