Eventuate Tram Customers and Orders

This application demonstrates two key patterns:

• Sagas - implement transactions that span services

• CQRS - implement queries that retrieve data from multiple services.

The application consists of three services:

• Order Service - manages orders

• Customer Service - manages customers

• Order History Service - maintains the order history

All services are implemented using Spring Boot, JPA and the Eventuate Tram framework, which provides transactional publish/subscribe.

The Order Service uses a choreography-based saga to enforce the customer’s credit limit when creating orders.

The Order History Service implements a CQRS view and subscribes to domain events published by the Order Service and Customer Service

About Sagas

Sagas are a mechanism for maintaining data consistency in a microservice architecture. A saga is a sequence of transactions, each of which is local to a service.

There are two main ways to coordinate sagas: orchestration and choreography. This example uses choreography-based sagas, which use domain events for coordination. Each step of a saga updates the local database and publishes a domain event. The domain event is processed by an event handler, which performs the next local transaction.

To learn more about why you need sagas if you are using microservices:

• Look at the Orchestration-based saga example

• Read the Saga pattern

• Look at MicroCPH 2019 presentation

• Read about sagas in my Microservices patterns book

The Create Order saga

The saga for creating an Order consists of the follow steps:

1. The Order Service creates an Order in a PENDING state and publishes an OrderCreated event

2. The Customer Service receives the event attempts to reserve credit for that Order. It publishes either a Credit Reserved event or a CreditLimitExceeded event.

3. The Order Service receives the event and changes the state of the order to either APPROVED or REJECTED.

About Command Query Responsibility Segregation (CQRS)

The CQRS pattern implements queries that retrieves data from multiple services. It maintains a queryable replica of the data by subscribing to domain events published by the services that own the data.

In this example, the Order History Service maintains a CQRS view in MongoDB by subscribing to domain events published by the Order Service and Customer Service. The CQRS view stores each customer as a MongoDB document that contains information the customer and their orders.

To learn more about why you need CQRS if you are using microservices:

• Read the CQRS pattern

• Look at GOTO Chicago 2019 presentation

• Read about CQRS in my Microservices patterns book

Transactional messaging with Eventuate Tram

The services uses the Eventuate Tram framework to communicate asynchronously using events. The flow for publishing a domain event using Eventuate Tram is as follows:

1. Eventuate Tram inserts events into the MESSAGE table as part of the ACID transaction that updates the JPA entity.

2. The Eventuate Tram CDC service tracks inserts into the MESSAGE table using the MySQL binlog (or Postgres WAL) and publishes messages to Apache Kafka.

3. A service subscribes to the events, updates its database, and possibly publishes more events.

Architecture

The following diagram shows the architecture of the Customers and Orders application.

The application consists of three services: Customer Service, Order Service, and Order History Service

Customer Service

The Customer Service implements a REST API for managing customers. The service persists the Customer JPA entity in a MySQL/Postgres database. Using Eventuate Tram, it publishes Customer domain events that are consumed by the Order Service.

For more information, see the microservice canvas for the Customer Service.

Order Service

The Order Service implements REST API for managing orders. The service persists the Order JPA entity in MySQL/Postgres database. Using Eventuate Tram, it publishes Order domain events that are consumed by the Customer Service.

For more information, see the microservice canvas for the Order Service.

Order History Service

The Order History Service implements REST API for querying a customer’s order history This service subscribes to events published by the Order Service and Customer Service and updates a MongoDB-based CQRS view.

For more information, see the microservice canvas for the Order History Service.

Building and running

Note: you do not need to install Gradle since it will be downloaded automatically. You just need to have Java 8 installed.

First, build the application

./gradlew assemble

Next, launch the services using Docker Compose:

./gradlew mysqlbinlogComposeBuild mysqlbinlogComposeUp

Note:

If the containers aren’t accessible via localhost - e.g. you are using Docker Toolbox, you will have to use ${DOCKER_HOST_IP} instead of localhost. See this guide to setting DOCKER_HOST_IP for more information.

You can also run the Postgres version using ./gradlew postgrespollingComposeUp or ./gradlew postgreswalComposeUp

Using the application

Once the application has started, you can use the application via the Swagger UI:

• Customer Service - http://localhost:8082/swagger-ui.html

• Order Service - http://localhost:8081/swagger-ui.html

• Order History Service - http://localhost:8083/swagger-ui.html

You can also use curl to interact with the services. First, let’s create a customer:

$ curl -X POST --header "Content-Type: application/json" -d '{
  "creditLimit": {
    "amount": 5
  },
  "name": "Jane Doe"
}' http://localhost:8082/customers

HTTP/1.1 200
Content-Type: application/json;charset=UTF-8

{
  "customerId": 1
}

Next, create an order:

$ curl -X POST --header "Content-Type: application/json" -d '{
  "customerId": 1,
  "orderTotal": {
    "amount": 4
  }
}' http://localhost:8081/orders

HTTP/1.1 200
Content-Type: application/json;charset=UTF-8

{
  "orderId": 1
}

Next, check the status of the Order in the Order Service:

$ curl -X GET http://localhost:8081/orders/1

HTTP/1.1 200
Content-Type: application/json;charset=UTF-8

{
  "orderId": 1,
  "orderState": "APPROVED"
}

Finally, look at the customer’s order history in the Order History Service:

$ curl -X GET --header "Accept: */*" "http://localhost:8083/customers/1"

HTTP/1.1 200
Content-Type: application/json;charset=UTF-8

{
  "id": 1,
  "orders": {
    "1": {
      "state": "APPROVED",
      "orderTotal": {
        "amount": 4
      }
    }
  },
  "name": "Chris",
  "creditLimit": {
    "amount": 100
  }
}

Got questions?

Don’t hesitate to create an issue or see

• Mailing list

• Slack Workspace

• Contact us.