Microservice architecture | message bus and driver - [bus & stream]

§1 The message bus

§1.1 brief introduction

The message bus

• In microservice system , Specify a global TOPIC
• this TOPIC Monitored and consumed by all instances of all services
• Be thrown into this TOPIC All the news in , Will be received by all services , So it's called The message bus

BUS

Spring Cloud Bus links nodes of a distributed system with a lightweight message broker. This can then be used to broadcast state changes (e.g. configuration changes) or other management instructions. AMQP and Kafka broker implementations are included with the project. Alternatively, any Spring Cloud Stream binder found on the classpath will work out of the box as a transport.

Spring Cloud Bus adopt Lightweight message brokers connect nodes in distributed systems . This can be used for Broadcast status changes （ such as , Configuration update ） Or others Management instructions . This project Include AMQP（Advanced Message Queuing Protocol） The protocol and kafka agent Realization . meanwhile , In the classpath （classpath） Anything you can find on Spring Cloud Stream binder Can be used as a transmission tool out of the box .

§1.2 Use

Get ready rabbitmq
Specific reference middleware | Message queue - [RabbitMq]

all Add the following content to the service that requires message queuing

rely on

<dependency>
    <groupId>org.springframework.cloud</groupId>
    <artifactId>spring-cloud-starter-bus-amqp</artifactId>
</dependency>

To configure

spring:
  rabbitmq:
    host: 192.168.3.7
    port: 5672
    username: guest
    password: guest

§2 Message driven

§2.1 brief introduction

Message driven
Be similar to JDBC , In complex architecture , Different subsystems are likely to use different message oriented middleware , At this time, message driven is often introduced , What works is

• Shield the underlying differences between different message oriented middleware
• Unified messaging programming model
• Reduce the switching cost of different message middleware and the learning cost of R & D personnel

Stream

Spring Cloud Stream is a framework for building highly scalable event-driven microservices connected with shared messaging systems.
The framework provides a flexible programming model built on already established and familiar Spring idioms and best practices, including support for persistent pub/sub semantics, consumer groups, and stateful partitions.

Spring Cloud Stream Is a Highly scalable microservices And Shared message system The connected framework .
This framework provides a flexible programming model , This model is based on the well-known spring Grammar and best practices , Including the release - Subscription persistence 、 Consumer groups and stateful partition Support for

binder
binder It is to access different message oriented middleware Stream The implementation of the , similar jdbc Connection pool and Driver The relationship between .
What we have now binder Realization ：

• RabbitMQ
• Apache Kafka
• Kafka Streams
• Amazon Kinesis
• Google PubSub (partner maintained)
• Solace PubSub+ (partner maintained)
• Azure Event Hubs (partner maintained)
• Azure Service Bus (partner maintained)
• AWS SQS (partner maintained)
• AWS SNS (partner maintained)
• Apache RocketMQ (partner maintained)

§2.2 principle

Statement
This post is essentially a note , The following principles do not involve specific code implementation , It's a set of things that the author rolls out to make himself understand
therefore , There is no guarantee that the understanding must be right （ Or there is almost certainly something wrong ）

Several important concepts
binder

• Binder , Used to create binding (binding) Tools for
• A binder , Basically, it can be understood as corresponding to a message oriented middleware , It can be analogized as right Database instance The connection of

binding

• binding , Binder (binder) Products
• The two ends of the binding are channel (MessageChannel) and Message exchange (Exchange)
• It can be analogized as right A specific database in the database instance The connection of , But because of binding one end channel (MessageChannel) It has its own direction , So it is only equivalent to the read or write part of a connection

MessageChannel

• channel , The channel of the message , Connecting business services and middleware , For a clear middleware , It acts as an agent (agent, No broker) The role of
• In business , Sending and receiving messages and other operations are actually realized through the operation channel ,Stream It realizes the decoupling of specific middleware through the channel

Source and @Output

• source coordination @EnableBinding Use
• source Marked the current service as a Message source To establish the binding (binding), The message source is the sender of the message 、 Exporter （@Output）
• source A channel is declared in , And mark it as @Output, Here's the picture
  
  Sink and @Input
• sink coordination @EnableBinding Use
• sink Marked the current service as a Message receiver To establish the binding (binding), The message receiver is the receiver of the message 、 Input by （@Input）
• sink A channel is declared in , And mark it as @Input

Schematic diagram ( With rabbit For example )

C Bit channel

• Stream Shielding the underlying differences of different message middleware is because it provides a set of general programming model
• The carrier of the general programming model is Message channel (MessageChannel)

The role of the channel

• The channel can be regarded as a set of standards or default implementation , Or simply understood as a standard API
• After the message channel appears , Researchers operate it , Realize the operation of message oriented middleware

How channels work

• In order to achieve the above purpose , The message channel should establish physical connection and logical binding with business services and message middleware at the same time
  • Connect to business services , Finally, you can use MessageChannel Of API Realize the proxy of message oriented middleware
  • Connect with message oriented middleware , Realize the physical input of messages / Output message middleware
• On the whole ： service –( Configure by annotation )– Message channel –( Access through binding connection )– Message middleware
  • Through annotation , Mark an input （@Input） Or output （@Output） Direction channel
  • Messages sent or monitored by the service , Will be transmitted to the channel first , The latter logic is represented by the channel
  • The channel will be configured according to the relevant , Create and specify message middleware instances that specify Exchange Binding relationship of
  • The channel realizes the message transmission with the message middleware through binding

§2.3 Use

rely on
According to the actual use of message oriented middleware xxx The content of

<dependency>
    <groupId>org.springframework.cloud</groupId>
    <artifactId>spring-cloud-starter-stream-xxx</artifactId>
</dependency>

To configure

spring:
  application:
    name: rabbit-stream-provider
  rabbitmq:
    host: 192.168.3.10
    port: 5672
    username: guest
    password: guest
  cloud:
      stream:
        binders: #  Configure the... To be bound here rabbitmq Service information ;
          defaultRabbit: #  Represents the name of the definition , Used on binding Integrate 
            type: rabbit #  Message component type 
        bindings: #  Integration of services 
          output: #  This name is the name of a channel 
            destination: studyExchange #  Indicates what to use Exchange Name definition 
            content-type: application/json #  Set message type , This time is json, Text is set to “text/plain”
            binder: defaultRabbit #  Set the specific settings of the message service to be bound 

§3 Message queuing common pitfalls

In the post next door