一、rabbitmq
RabbitMQ是一个在AMQP基础上完整的,可复用的企业消息系统。他遵循Mozilla Public License开源协议。
MQ全称为Message Queue, 消息队列(MQ)是一种应用程序对应用程序的通信方法。应用程序通过读写出入队列的消息(针对应用程序的数据)来通信,而无需专用连接来链接它们。消 息传递指的是程序之间通过在消息中发送数据进行通信,而不是通过直接调用彼此来通信,直接调用通常是用于诸如远程过程调用的技术。排队指的是应用程序通过 队列来通信。队列的使用除去了接收和发送应用程序同时执行的要求。
1.1 安装rabbitmq
RabbitMQ安装
1
2
3
4
5
6
7
8
|
安装配置epel源 $ rpm - ivh http: / / dl.fedoraproject.org / pub / epel / 6 / i386 / epel - release - 6 - 8.noarch .rpm 安装erlang $ yum - y install erlang 安装RabbitMQ $ yum - y install rabbitmq - server |
注意:service rabbitmq-server start/stop
安装API:
1
2
3
4
5
6
7
8
9
|
pip install pika or easy_install pika or 源码 or pycharm https: / / pypi.python.org / pypi / pika |
1.3 用python操作rabbitmq
1.3.1 基于Queue实现生产者消费者模型
#!/usr/bin/env python # -*- coding:utf-8 -*- import Queue import threading message = Queue.Queue(10) def producer(i): while True: message.put(i) def consumer(i): while True: msg = message.get() for i in range(12): t = threading.Thread(target=producer, args=(i,)) t.start() for i in range(10): t = threading.Thread(target=consumer, args=(i,)) t.start()
1.3.2 rabbitmq实现消息队列
对于RabbitMQ来说,生产和消费不再针对内存里的一个Queue对象,而是某台服务器上的RabbitMQ Server实现的消息队列。
先运行消费者脚本,让它监听队列消息,然后运行生产者脚本,生产者往队列里发消息。然后消费者往队列里取消息。
import pika # ########################### 消费者 ########################### connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208')) channel = connection.channel() channel.queue_declare(queue='abc') # 如果队列没有创建,就创建这个队列 def callback(ch, method, propertities,body): print(" [x] Received %r" % body) channel.basic_consume(callback, queue='abc', # 队列名 no_ack=True) # 不通知已经收到,如果连接中断可能消息丢失 print(' [*] Waiting for message. To exit press CTRL+C') channel.start_consuming()
import pika # ############################## 生产者 ############################## connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208' )) channel = connection.channel() channel.queue_declare(queue='abc') # 如果队列没有创建,就创建这个队列 channel.basic_publish(exchange='', routing_key='abc', # 指定队列的关键字为,这里是队列的名字 body='Hello World!') # 往队列里发的消息内容 print(" [x] Sent 'Hello World!'") connection.close()
先运行消费者,然后再运行生产者:
''' 打印: 生产者: [x] Sent 'Hello World!' 消费者: [*] Waiting for message. To exit press CTRL+C [x] Received b'Hello World!' '''
1.4 no-ack=False:rabbitmq消费者连接断了 消息不丢失
rabbitmq支持一种方式:应答。比如我从消息里拿一条消息,如果全处理完,你就不要帮我记着了。如果没处理完,突然断开了,再连接上的时候,消息队列就会重新发消息。
总结:
- Basic.Ack 发回给 RabbitMQ 以告知,可以将相应 message 从 RabbitMQ 的消息缓存中移除。
- Basic.Ack 未被 consumer 发回给 RabbitMQ 前出现了异常,RabbitMQ 发现与该 consumer 对应的连接被断开,之后将该 message 以轮询方式发送给其他 consumer (假设存在多个 consumer 订阅同一个 queue)。
- 在 no_ack=true 的情况下,RabbitMQ 认为 message 一旦被 deliver 出去了,就已被确认了,所以会立即将缓存中的 message 删除。所以在 consumer 异常时会导致消息丢失。
- 来自 consumer 侧的 Basic.Ack 与 发送给 Producer 侧的 Basic.Ack 没有直接关系
注意:
1)只有在Consumer(消费者)断开连接时,RabbitMQ才会重新发送未经确认的消息。
2)超时的情况并未考虑:无论Consumer需要处理多长时间,RabbitMQ都不会重发消息。
消息不丢失的关键代码:
1)在接收端的callback最后:
channel.basic_ack(delivery_tag=method.delivery_tag)
1
2
3
|
ack即acknowledge(承认,告知已收到) 也就是消费者每次收到消息,要通知一声:已经收到,如果消费者连接断了,rabbitmq会重新把消息放到队列里,下次消费者可以连接的时候,就能重新收到丢失消息。 A message MUST not be acknowledged morethan once. The receiving peer MUST validate that a non - zero delivery - tag refersto a delivered message, <br> and raise a channel exception if this is not the case. |
2)除了callback函数,还要在之前设置接收消息时指定no_ack(默认False):
channel.basic_consume(callback, queue='hello', no_ack=False)
消费者:
import pika
# ########################### 消费者 ##########################
connection = pika.BlockingConnection(pika.ConnectionParameters( host='10.211.55.4')) channel = connection.channel() channel.queue_declare(queue='hello') def callback(ch, method, properties, body): print(" [x] Received %r" % body) import time time.sleep(10) print('ok') ch.basic_ack(delivery_tag = method.delivery_tag) channel.basic_consume(callback, queue='hello', no_ack=False) print(' [*] Waiting for messages. To exit press CTRL+C') channel.start_consuming()
消费者断掉连接,再次连接,消息还会收到。
1.5 durable:rabbitmq服务端宕机 消息不丢失
发数据的时候,就说了:我这条数据要持久化保存。
如果rabbitmq服务端机器如果挂掉了,会给这台机器做持久化。如果启动机器后,消息队列还在。
生产者.py:
import pika
# ############################## 生产者 ##############################
connection = pika.BlockingConnection(pika.ConnectionParameters(host='10.211.55.4')) channel = connection.channel() # make message persistent channel.queue_declare(queue='hello', durable=True) channel.basic_publish(exchange='', routing_key='hello', body='Hello World!', properties=pika.BasicProperties( delivery_mode=2, # make message persistent )) print(" [x] Sent 'Hello World!'") connection.close()
消费者.py:
import pika
# ########################### 消费者 ###########################
connection = pika.BlockingConnection(pika.ConnectionParameters(host='10.211.55.4')) channel = connection.channel() # make message persistent channel.queue_declare(queue='hello', durable=True) def callback(ch, method, properties, body): print(" [x] Received %r" % body) import time time.sleep(10) print('ok') ch.basic_ack(delivery_tag = method.delivery_tag) channel.basic_consume(callback, queue='hello', no_ack=False) print(' [*] Waiting for messages. To exit press CTRL+C') channel.start_consuming()
测试:
1)把生产者.py执行三次。
2)然后在linux上停掉rabbitmq服务,然后再开启rabbitmq服务
1
2
3
4
5
6
|
[root@localhost ~] # /etc/init.d/rabbitmq-server stop Stopping rabbitmq - server: rabbitmq - server. [root@localhost ~] # /etc/init.d/rabbitmq-server start Starting rabbitmq - server: SUCCESS rabbitmq - server. |
3)运行:消费者.py:三条消息都打印了:
1
2
3
4
5
6
7
|
[ * ] Waiting for messages. To exit press CTRL + C [x] Received b 'Hello World!' ok [x] Received b 'Hello World!' ok [x] Received b 'Hello World!' ok |
1.6 消息获取顺序
默认消息队列里的数据是按照顺序被消费者拿走,例如:消费者1 去队列中获取 奇数 序列的任务,消费者1去队列中获取 偶数 序列的任务。
因为默认是跳着取得。第一个消费者取得很快,已经执行到20了,但是第二个消费者只取到13,可能消息执行的顺序就有问题了。
如果多个消费者,如果不想跳着取,就按消息的顺序取,而不是按着自己的间隔了。
channel.basic_qos(prefetch_count=1) 表示谁来谁取,不再按照奇偶数排列
#!/usr/bin/env python # -*- coding:utf-8 -*- __author__ = 'WangQiaomei' import pika # ########################### 消费者 ########################### connection = pika.BlockingConnection(pika.ConnectionParameters(host='192.168.137.208')) channel = connection.channel() # make message persistent channel.queue_declare(queue='hello1') def callback(ch, method, properties, body): print(" [x] Received %r" % body) import time time.sleep(10) print('ok') ch.basic_ack(delivery_tag = method.delivery_tag) channel.basic_qos(prefetch_count=1) channel.basic_consume(callback, queue='hello1', no_ack=False) print(' [*] Waiting for messages. To exit press CTRL+C') channel.start_consuming()
1.7发布订阅
发布订阅原理:
1)发布订阅和简单的消息队列区别在于,发布订阅会将消息发送给所有的订阅者,而消息队列中的数据被消费一次便消失。
2)所以,RabbitMQ实现发布和订阅时,会为每一个订阅者创建一个队列,而发布者发布消息时,会将消息放置在所有相关队列中。
3)exchange 可以帮你发消息到多个队列!type设为什么值,就把消息发给哪些队列。
发布订阅应用到监控上:
模板就是写上一段脚本,放在服务器上,
客户端每5分钟,从服务端拿到监控模板,根据模板来取数据,
然后把数据结果发步到服务端的redis频道里。
服务端收到数据,1)处理历史记录 2)报警 3)dashboard显示监控信息
服务端有三处一直来订阅服务端频道(一直来收取客户端监控数据)
1.7.1 发布给所有绑定队列
exchange type = fanout
exchange 可以帮你发消息到多个队列,type = fanout表示:跟exchange绑定的所有队列,都会收到消息。
发布者:
#!/usr/bin/env python # -*- coding: utf-8 -*- __author__ = 'WangQiaomei import pika import sys # ########################### 发布者 ########################### connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208')) channel = connection.channel() channel.exchange_declare(exchange='logs', type='fanout') message = ' '.join(sys.argv[1:]) or "info: Hello World!" channel.basic_publish(exchange='logs', routing_key='', body=message) print(" [x] Sent %r" % message) connection.close()
订阅者:
#!/usr/bin/env python # -*- coding: utf-8 -*- __author__ = 'WangQiaomei' import pika # ########################### 订阅者 ########################### connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208')) channel = connection.channel() channel.exchange_declare(exchange='logs', type='fanout') # 随机创建队列 result = channel.queue_declare(exclusive=True) queue_name = result.method.queue # 绑定 channel.queue_bind(exchange='logs', 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() ''' 多次执行这个文件,就会随机生成多个队列。并且exchange都绑定这些队列。 然后发布者只需要给exchange发送消息,然后exchange绑定的多个队列都有这个消息了。订阅者就收到这个消息了。 '''
1.7.2关键字发送
一个队列还可以绑定多个关键字
对一个随机队列,绑定三个关键字
再次执行,对另一个随机队列,只绑定一个关键字。
消费者:每执行一次可以生成一个队列。通过使用命令行传参的方式,来传入队列的关键字。
#!/usr/bin/env python 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 severities :] if not severities: sys.stderr.write("Usage: %s [info] [warning] [error] " % sys.argv[0]) sys.exit(1) 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()
容易测试的版本:
消费者1:
#!/usr/bin/env python # -*- coding: utf-8 -*- __author__ = 'WangQiaomei' import pika import sys # ########################### 消费者1 ########################### connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208')) channel = connection.channel() channel.exchange_declare(exchange='direct_logs', type='direct') result = channel.queue_declare(exclusive=True) # 随机生成队列 queue_name = result.method.queue severities ] if not severities: sys.stderr.write("Usage: %s [info] [warning] [error] " % sys.argv[0]) sys.exit(1) 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()
消费者2:
#!/usr/bin/env python # -*- coding: utf-8 -*- __author__ = 'WangQiaomei' import pika import sys # ########################### 消费者2 ########################### connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208')) channel = connection.channel() channel.exchange_declare(exchange='direct_logs', type='direct') result = channel.queue_declare(exclusive=True) # 随机生成队列 queue_name = result.method.queue severities ] 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()
生产者:
#!/usr/bin/env python # -*- coding: utf-8 -*- __author__ = 'WangQiaomei' import pika import sys # ############################## 生产者 ############################## connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208')) channel = connection.channel() channel.exchange_declare(exchange='direct_logs', type='direct') severity = 'info' message = 'Hello World!' channel.basic_publish(exchange='direct_logs', routing_key=severity, body=message) print(" [x] Sent %r:%r" % (severity, message)) connection.close() ''' 同时运行消费者1,消费者2,然后修改生产者的关键字,运行生产者。 当生产者:severity = 'info',则消费者1收到消息,消费者2没收到消息 当生产者:severity = 'error',则消费者1、消费者2 都收到消息 '''
1.7.2 模糊匹配
exchange type = topic
在topic类型下,可以让队列绑定几个模糊的关键字,之后发送者将数据发送到exchange,exchange将传入”路由值“和 ”关键字“进行匹配,匹配成功,则将数据发送到指定队列。
- # 表示可以匹配 0 个 或 多个 字符
- * 表示只能匹配 一个 任意字符
1
2
3
|
发送者路由值 队列中 old.boy.python old. * - - 不匹配 old.boy.python old. # -- 匹配 |
消费者:
#!/usr/bin/env python import pika import sys # ############################## 消费者 ############################## connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208')) channel = connection.channel() channel.exchange_declare(exchange='topic_logs', type='topic') result = channel.queue_declare(exclusive=True) queue_name = result.method.queue binding_keys = "*.orange.*" for binding_key in binding_keys: channel.queue_bind(exchange='topic_logs', queue=queue_name, routing_key=binding_key) 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()
生产者:
#!/usr/bin/env python import pika import sys # ############################## 生产者 ############################## connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208')) channel = connection.channel() channel.exchange_declare(exchange='topic_logs', type='topic') # routing_key = 'abc.new.qiaomei.old' routing_key = 'neworangeold' message = 'Hello World!' channel.basic_publish(exchange='topic_logs', routing_key=routing_key, body=message) print(" [x] Sent %r:%r" % (routing_key, message)) connection.close() ''' #.orange.# 匹配:new.orange.old neworangeold *.orange.* 匹配:neworangeold,不匹配:new.orange.old '''
1.8 saltstack原理实现
saltstack:zeromq:放到内存里的,会更快,会基于这个做rcp
openstack:大量使用:rabbitmq
saltstack上有master,有三个队列。,让三个客户端每个人取一个队列的任务
saltstack的原理:
1)发一条命令ifconfig,想让所有nginx主机组的机器,都执行。
2)在master我们可以发命令给exchange,nginx总共有10台服务器,创建10个带有nginx关键字的10个队列,
3)master随机生成队列,md5是一个队列的名字,exchange把命令和md5这个消息推送到nginx关键字的队列里。
4)nginx10台服务器从队列中取出消息,执行命令,并且把主机名和执行的结果返回给这个队列里。
5)master变为消费者,取出队列里的主机名和执行结果,并打印到终端上。
服务器1:
#!/usr/bin/env python # -*- coding: utf-8 -*- __author__ = 'WangQiaomei' import pika import sys # ########################### 消费者1 ########################### connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208')) channel = connection.channel() channel.exchange_declare(exchange='direct_logs', type='direct') result = channel.queue_declare(exclusive=True) # 随机生成队列 queue_name = result.method.queue severities = ] if not severities: sys.stderr.write("Usage: %s [info] [warning] [error] " % sys.argv[0]) sys.exit(1) 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)) queue_md5=body.decode().split(",")[1] hostname = 'nginx1' channel.queue_declare(queue=queue_md5) # 如果队列没有创建,就创建这个队列 channel.basic_publish(exchange='', routing_key=queue_md5, # 指定队列的关键字为,这里是队列的名字 body='%s|cmd_result1' %hostname) # 往队列里发的消息内容 channel.basic_consume(callback, queue=queue_name, no_ack=True) channel.start_consuming()
服务器2:
#!/usr/bin/env python # -*- coding: utf-8 -*- __author__ = 'WangQiaomei' import pika import sys # ########################### 消费者2 ########################### connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208')) channel = connection.channel() channel.exchange_declare(exchange='direct_logs', type='direct') result = channel.queue_declare(exclusive=True) # 随机生成队列 queue_name = result.method.queue severities = ["nginx"] 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)) queue_md5=body.decode().split(",")[1] hostname = 'nginx2' channel.queue_declare(queue=queue_md5) # 如果队列没有创建,就创建这个队列 channel.basic_publish(exchange='', routing_key=queue_md5, # 指定队列的关键字为,这里是队列的名字 body='%s|cmd_result2' %hostname) # 往队列里发的消息内容 channel.basic_consume(callback, queue=queue_name, no_ack=True) channel.start_consuming()
master:
#!/usr/bin/env python # -*- coding: utf-8 -*- __author__ = 'WangQiaomei' import pika import sys import hashlib # ############################## 生产者 ############################## connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208')) channel = connection.channel() channel.exchange_declare(exchange='direct_logs', type='direct') severity = 'nginx' m2 = hashlib.md5() m2.update(severity.encode('utf-8')) md5_security=m2.hexdigest() print('md5_security:',md5_security) message = 'cmd,%s' % md5_security channel.basic_publish(exchange='direct_logs', routing_key=severity, body=message) print(" [x] Sent %r:%r" % (severity, message)) connection.close() #################################3 connection = pika.BlockingConnection(pika.ConnectionParameters( host='192.168.137.208')) channel = connection.channel() channel.queue_declare(queue=md5_security) # 如果队列没有创建,就创建这个队列 def callback(ch, method, propertities,body): print(" [x] Received %r" % body) channel.basic_consume(callback, queue=md5_security, # 队列名 no_ack=True) # 不通知已经收到,如果连接中断消息就丢失 print(' [*] Waiting for message. To exit press CTRL+C') channel.start_consuming()
打印:
1
2
3
4
5
6
7
8
9
10
11
12
|
''' 服务器1,和服务器2都打印: [*] Waiting for logs. To exit press CTRL+C [x] 'nginx':b'cmd,ee434023cf89d7dfb21f63d64f0f9d74' master打印: md5_security: ee434023cf89d7dfb21f63d64f0f9d74 [x] Sent 'nginx':'cmd,ee434023cf89d7dfb21f63d64f0f9d74' [*] Waiting for message. To exit press CTRL+C [x] Received b'nginx2|cmd_result2' [x] Received b'nginx1|cmd_result1' ''' |