Running Stork Locally

[hablocher]

Folks, please, i need some help!

I want to use Stork in a standalone application, whitout Spring or Quarkus. For example, i want to use Consul as my Service Discovery. But i can´t find any documentation which can clarify to me how to do this.

Which dependencies i need to declare in pom.xml? Where i need to put configuration? How Stork find the properties configuration, for example, stork.my-service.service-discovery.type=consul?

In the following snippet

public static void main(String[] args) {
Stork.initialize();
Stork stork = Stork.getInstance();
// ...
}
how Stork find the configuration for Consul, for example, and consequently finds the registered services?

[cescoffier]

It depends.
If you are in a CDI environment, I recommend using a Microprofile config implementation (such as SmallRye Config). You must use the API to define your services if you are not in a CDI environment. See https://smallrye.io/smallrye-stork/2.3.2/programmatic-api/ for details.

[hablocher]

I found the hard way to run standalone!

• Put the service discovery dependency (for example, stork-service-discovery-consul) in pom.xml
• Put the dependency stork-configuration-generator in pom.xml to generate io.smallrye.stork.api.config.ConfigWithType implementation based in @ServiceDiscoveryAttribute and @ServiceDiscoveryType annotations and the io.smallrye.stork.spi.ServiceDiscoveryProvider interface (which are implemented in stork-service-discovery-consul dependency, for example).
• Implement the interface io.smallrye.stork.spi.config.ConfigProvider to define the services and the values of parameters (normally, in CDI context, these values are located in application.properties file) and configure it to be finded using SPI

Folks, little and innocent feedback: your documentation suffer from some lack of information.

Thx! Cya!

[cescoffier]

It seems you decided to build your own provider. This is not what your asked before (we have a different documentation for that use case)

[cescoffier]

So, here is a complete example, using the static list discovery, but you can use any discovery/selection:

First, the pom.xml file:

<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0"
         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
    <modelVersion>4.0.0</modelVersion>

    <groupId>org.example</groupId>
    <artifactId>stork-bare</artifactId>
    <version>1.0-SNAPSHOT</version>

    <properties>
        <maven.compiler.source>17</maven.compiler.source>
        <maven.compiler.target>17</maven.compiler.target>
        <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
    </properties>

    <dependencies>
        <dependency>
            <groupId>io.smallrye.stork</groupId>
            <artifactId>stork-core</artifactId>
            <version>2.3.2</version>
        </dependency>
        <dependency>
            <groupId>io.smallrye.stork</groupId>
            <artifactId>stork-service-discovery-static-list</artifactId>
            <version>2.3.2</version>
        </dependency>
    </dependencies>

</project>

Then, the code:

package me.escoffier.stork.demo;

import io.smallrye.stork.Stork;
import io.smallrye.stork.api.ServiceDefinition;
import io.smallrye.stork.servicediscovery.staticlist.StaticConfiguration;

public class Main {

	public static void main(String[] args) {
		Stork.initialize();
		var stork = Stork.getInstance()
				.defineIfAbsent("my-service", ServiceDefinition.of(new StaticConfiguration()
						.withAddressList("localhost:9000, localhost:9001")
						.withShuffle("true"))); // Using the round-robin, but you can also configure a load balancer too.

		var svc = stork.getService("my-service");
		for (int i = 0; i < 5; i++) {
			var instance = svc.selectInstance().await().indefinitely();
			System.out.printf("Selected instance is : %d, %s:%d%n", instance.getId(), instance.getHost(), instance.getPort());
		}

	}
}

So, I don't understand how you went on the annotation processing path.

[hablocher]

You are right! Take some time to me understand the concepts behind stork. What i needed it was a way to register a service on the fly, dynamically.
In my study, i found the annotation processor path... but using it i had some fellings that i was in the wrong way :P
Your comment clarified all the things... and, to be fair, i found this sample in the documentation... sorry for say the documentation was suffering from lack of infomation :(

Thanks a lot for your reply!