Spring Cloud Stream

RabbitMQ

Run broker:

docker run --rm -it --name rabbitmq -p 5672:5672 -p 15672:15672 rabbitmq:management

Projects

• rabbitmq-events-producer:
  • produces the messages and expose a REST API to publish a message
  • has two suppliers: one to simple message and another with routing key
• rabbitmq-events-consumer: show how to build a simple consumer
• rabbitmq-events-consumer-routing: show how to use routing expression with differents consumers
• rabbitmq-events-partitioning: show how to use partitioned consumers
• rabbitmq-events-consumer-testing: show how to test a consumer

Bellow is the complete diagram:

Producer

The producer has two supplier that can be defined using the property spring.cloud.function.name.

The producer sends a message each second with a header to_process=true.

The posting a message using REST API, it will be sent without any header so the routing consumer can only log it.

POST localhost:8080/security/events

{
	"message": "Message to be logged"
}

Consumers

There are 3 consumers:

• simple consumer: only logs the message (pritns Security event consumed);
• routing: defines two functions that is routed depending on header to_process=true
  • no header: logs the message (prints Security event logged);
  • header: logs the message (prints Security event processed) and return a value to another Exchange.
• partitioning: logs the message (prints Security event processed from consumer index #) from its specified consumer index (env spring.cloud.stream.instanceIndex).

To run the partitioned consumer we can set the INSTANCE_INDEX with the consumer index:

export INSTANCE_INDEX=0
mvn spring-boot:run

Bellow is the bindings running two instances:

Kafka

Run broker:

docker-compose up

Projects

• kafka-events-producer: show how to send message
• kafka-events-consumer: show how to build a simple topic consumer
• kafka-events-joiner-consumer:
  • show how to consume from two topics and produce the result to another one
  • show how to store aggregation result into a persistent store
• kafka-events-consumer-error-handling: show how to different error handling and how to setup a DLQ
• kafka-stream-functions-samples: show different cases of streaming

Bellow is the complete diagram:

References

• Spring Cloud Stream - RabbitMQ