This is a demo trading application. It's a simple Java/Spring Boot microservice, for buying and selling stocks.

(Note that it doesn't really buy or sell anything, it's just a simulation.)

The application is designed to show how to add observability to a Java app, so it's configured with a couple of extra bits to make it easier to observe:

• Metrics. Using the open source library Micrometer, the application exposes some metrics at a Prometheus-compatible endpoint (/actuator/prometheus).

• Logs. The application writes logs to the console for each trade, which can be scraped and sent to a log aggregator like Promtail/Loki.

To run in Kubernetes, you will need:

• A Kubernetes cluster

• Grafana Agent installed in your Kubernetes cluster, configured to push metrics to your Grafana Cloud account, OR Prometheus installed in your Kubernetes cluster, if you're not using Grafana Cloud.

First we need to configure Prometheus or Grafana Agent to collect metrics from the application. This can be done in lots of different ways, but we're using annotations.

We've added these annotations to our Kubernetes Service:

prometheus.io/scrape: "true"
prometheus.io/port: "8080"
prometheus.io/path: "/actuator/prometheus"

These Kubernetes annotations are a fairly common way to advertise to a capable scraper that this application exposes metrics, and we want them to be scraped.

To make sure that our microservice's metrics get picked up by Grafana Agent or Prometheus, ensure that your Agent/Prometheus has a job to scrape annotated Kubernetes endpoints. The configuration for the job should look something like this:

configs:
- name: your-config-name
    ...
    scrape_configs:
    ...
    - job_name: integrations/kubernetes/pod-app-metrics
        kubernetes_sd_configs:
        - role: endpoints
        relabel_configs:
        - source_labels: [__meta_kubernetes_service_annotation_prometheus_io_scrape]
            action: keep
            regex: true
        - source_labels: [__meta_kubernetes_service_annotation_prometheus_io_path]
            action: replace
            target_label: __metrics_path__
            regex: (.+)
        - source_labels: [__address__, __meta_kubernetes_pod_annotation_prometheus_io_port]
            action: replace
            regex: ([^:]+)(?::\d+)?;(\d+)
            replacement: $1:$2
            target_label: __address__

Restart your Agent or Prometheus after making any changes. Then, Grafana Agent (or Prometheus) will automatically scrape the metrics from any Services with these prometheus.io/* annotations.

You will need to have the following installed on your machine:

• Java (if running standalone)
• Docker (if running the complete application + log collector in a container)

To run the microservice on its own, you can run the following command:

cd trading-service
./mvnw spring-boot:run

There aren't many tests at the moment 😓, but to run them, you can run the following command:

cd trading-service
./mvnw test

Running a load test with k6.

First, install k6.

Then:

cd load-testing
npm install
npm run pretest
npm run test

This will run a short load test against the application, and output the results to the console.

• Maven - The build tool used by all the coolest Java developers (and some of the not-so-cool ones)
• Spring Boot - Hugely popular application framework for Java
• Micrometer - Application metrics library for Java that's vendor-neutral (which means you can use it to share data with Prometheus, InfluxDB, Datadog, etc.)
• Prometheus - Time-series database for gathering and storing metrics.
• k6 - Load testing tool that we're slightly misusing to simulate traffic to the application.

• Tom Donohue - monodot

Pssst. Some of this text was written with GitHub Copilot. 🤖 (Bet you can't tell which bits!)