Skip to content

0 · Setup

The workshop project lives in the EventConductor repository and builds against the engine from source:

Terminal window
git clone https://github.com/miguelperezcolom/eventconductor.git
cd eventconductor
mvn -B install -DskipTests -Djacoco.skip=true # installs the engine into ~/.m2 (a few minutes)
cd examples/workshop-hotel-booking
mvn verify # validates the definitions and runs the tests
Terminal window
mvn spring-boot:run

When the log says Started HotelBookingApplication, open http://localhost:8095.

The console: the Workflow menu over the list of definitions

The console lists the six workflow definitions of the workshop. The Workflow menu leads to everything you will use: Processes (the runs), Steps, Tasks, Incidents, Batch operations and more.

One Spring Boot application holds all of it — the point of this setup is that there is nothing else to install:

@WorkflowEmbeddedApplication // the engine, embedded in this app
@EnableWorkflowUi // its console
@EnableJpaRepositories(basePackages = "io.mateu.workflow.infra.out.persistence")
@AutoConfigurationPackage(basePackages = "io.mateu.workflow.infra.out.persistence")
public class HotelBookingApplication {
public static void main(String[] args) {
SpringApplication.run(HotelBookingApplication.class, args);
}
}

The essential configuration, from application.properties:

workflow.mode=embedded # the engine runs inside this application
workflow.persistence=jpa # the console reads the engine's tables…
spring.datasource.url=jdbc:h2:mem:hotel;DB_CLOSE_DELAY=-1 # …kept in an in-memory H2
# Workshop only: no login. Since 2.27/2.28 the REST API, the console and MCP refuse anonymous
# callers; this lets a laptop use them. NEVER in anything reachable from a network.
workflow.message-api.allow-anonymous=true

The data lives in memory: every restart starts clean. In production, the same engine runs on PostgreSQL, and — when one node is not enough — distributed over Kafka, with the same definitions: see Deployment Modes.