spring-cloud-sleuth/spring-cloud-sleuth-zipkin-stream

Running Zipkin Server

There are 3 parts to Zipkin: the instrumented client apps, the backend database and the Zipkin server. The database for this implementation is MySQL.

There is a running instance on PWS: http://zipkin-web.cfapps.io. It is backed by a zipkin-server with a MySQL backend and RabbitMQ (Spring Cloud Stream) for span transport.

Instrumenting Apps

Depend on Spring Cloud Sleuth Stream. Bind to a rabbit service (or redis if you prefer - normal Spring Cloud Stream process).

Zipkin Server

Depend on spring-cloud-sleuth-zipkin-stream and enable the server:

@SpringBootApplication
@EnableZipkinStreamServer
public class ZipkinStreamServerApplication {

	public static void main(String[] args) throws Exception {
		SpringApplication.run(ZipkinStreamServerApplication.class, args);
	}

}

Zipkin has a web UI, which is enabled by default when you depend on io.zipkin.java:zipkin-autoconfigure-ui.

Bind to MySQL and the same Stream service that you did in the apps (rabbit, redis, kafka). Set spring.datasource.initialize=true the first time you start to initialize the database.

Uses the zipkin-server library from the OSS as well as spring-cloud-sleuth-stream.

NOTE: running in the "test" profile you don't need MySQL (the span store is in memory). You could even run in PWS without MySQL.