Before we start to work, we need to configure a service which the Zipkin server needs. By default, the Zipkin server uses in-memory databases, but it is not recommended for production; usually, developers use this feature to demonstrate Zipkin features.
We will use MySQL as a data store. The Zipkin server also supports different sources, such as Cassandra and Elasticsearch.
Spring Cloud Sleuth supports synchronous and asynchronous operations. The synchronous operations are over the HTTP protocol and asynchronous can be done by RabbitMQ or Apache Kafka.
To use the HTTP, that is, REST API, we should use @EnableZipkinServer, it will delegate the persistence for REST tier through the SpanStore interface.
