NATS Proof of concept

This simple gradle project presents a proof of concept of microservices communicating through a NATS server.

Every instance of this application can be configured to and/or not to:

• Start a scheduler which, with a fixed rate time, triggers the doStuff method of the application service
• Start a NATS dispatcher which listens (in a queue way) for one topic and, every time it gets a message from that topic, triggers the doStuff method of the application service

At the moment, every time the doStuff method of the application service gets executed:

• It may send (50% chance) a nats message to one of the configured topics
• It stores the message in the mongo repository

This application can be run as standalone, but the idea is to deploy several replicas for each topic and create a network of instances of the same application. For that use case, take a look at the k8s README.

You can take a look at the roadmap file for my next steps with this project.

Compilation

This project is built with gradle. It is recommended to use the gradle wrapper provided with the project.

JDK 11 is needed.

For compiling the project, you can use this:

./gradlew clean build

Before Execution

Mongo requirement

If you do not have a mongo server running you can easily run it with docker with this:

docker run --memory=512m --name mongo \
    -p 27017:27017 \
    --restart unless-stopped \
    -d mongo

These are the standard mongodb properties to edit in case of need:

spring.data.mongodb.database=nats-poc
spring.data.mongodb.host=localhost
spring.data.mongodb.port=27017

Nats requirement

If you do not have a NATS server running you can easily run it with docker with this:

docker run --memory=512m --name nats \
    -p 4222:4222 \
    -p 6222:6222 \
    -p 8222:8222 \
    --restart=unless-stopped \
    -d nats -DV

Execution

./gradlew bootRun

Once it is running

Once the server is running (with the default properties) you can:

• Take a look at the application logs and the ingoing and outgoing communications
• Take a look at the nats server log
• Check the stored documents at the MongoDB