README.md

How to use bare Kafka clients in Quarkus

This project illustrates how to use bare Apache Kafka clients and Vert.x Kafka clients in a Quarkus application.

Kafka cluster

First, you need a Kafka cluster. You can follow the instructions from the Apache Kafka web site or run docker-compose up if you have docker installed on your machine.

Start the application

Start the application in dev mode using:

mvn quarkus:dev

Post a Kafka record

# post a record using the bare Kafka client
> http POST :8080/kafka key==bare value==value
# post a record using the Vert.x Kafka client
> http POST :8080/vertx-kafka key==bare value==value

Retrieve the last record

# retrieve a record using the bare Kafka client
> http :8080/kafka 

# retrieve a record using the Vert.x Kafka client
> http :8080/vertx-kafka

List existing topics

# retrieve the list of topics using the bare Kafka client
> http :8080/kafka/topics

# retrieve the list of topics using the Vert.x Kafka client
> http :8080/vertx-kafka/topics

Anatomy

The org.acme.kafka.KafkaProviders and org.acme.kafka.VertxKafkaProviders produces the Kafka consumers, produces and admin for the bare API and Vert.x API. The org.acme.kafka.KafkaEndpoint and org.acme.kafka.VertxKafkaEndpoint demonstrates how these clients can be used.

Note that the configuration is retrieved using:

@Inject
@Identifier("default-kafka-broker")
Map<String, Object> config;

This config contains all the kafka. properties from the application configuration.

Running in native

You can compile the application into a native binary using:

mvn clean install -Pnative

and run with:

./target/kafka-bare-quickstart-1.0-SNAPSHOT-runner