การผสานรวมกับ Prometheus และ Grafana

In modern systems, understanding what your applications and infrastructure are doing is critical. This is where monitoring comes in!

We'll explore how Prometheus and Grafana form a powerful duo for observing your Spring Boot Kafka applications and the Kafka cluster itself.

Prometheus is an open-source monitoring system that collects metrics from configured targets at given intervals, evaluates rule expressions, displays the results, and can trigger alerts.

• It uses a pull model: Prometheus scrapes (pulls) metrics from your services.
• It stores data as time-series data, making it excellent for tracking changes over time.

Spring Boot applications can expose metrics using Actuator, Spring's production-ready features. Micrometer is an instrumentation library that Actuator uses to provide a vendor-neutral metrics facade.

This allows you to export metrics to various monitoring systems, including Prometheus, with minimal code changes.

To make your Spring Boot app's metrics available for Prometheus, you need to add dependencies and configure Actuator.

First, add spring-boot-starter-actuator and micrometer-registry-prometheus to your pom.xml.

Then, enable the Prometheus endpoint in application.properties:

This exposes metrics at /actuator/prometheus.

This minimal Spring Boot application demonstrates how to expose a custom counter. When you hit the /hello endpoint, the app.hello.requests counter increments, which Prometheus can then scrape.

Kafka brokers don't natively expose metrics in a Prometheus-friendly format. We use a separate component called Kafka Exporter.

Kafka Exporter is a lightweight service that scrapes JMX metrics from Kafka brokers and re-exposes them in a format that Prometheus can easily consume.

Prometheus needs to know where to find your metrics. You configure scrape targets in its prometheus.yml file. Here's an example:

Grafana is an open-source platform for monitoring and observability. It allows you to query, visualize, alert on, and understand your metrics no matter where they are stored.

Think of it as the display panel for all the data Prometheus collects. It uses dashboards with various panels to present metrics in an easy-to-understand way.

Before visualizing, Grafana needs to connect to a data source. Here, Prometheus will be our data source.

In Grafana's UI, navigate to "Connections" -> "Data sources" -> "Add new data source". Select "Prometheus" and configure its URL (e.g., http://localhost:9090, Prometheus' default port).

With Prometheus connected, you can start building dashboards.

• Create a new dashboard and add a panel.
• Select Prometheus as the data source.
• Write a PromQL query (Prometheus Query Language) to select your metric, e.g., app_hello_requests_total for our Spring Boot app.
• Choose a visualization type like "Graph" or "Stat".

This brings your metrics to life!

You've learned how Prometheus scrapes metrics and Grafana visualizes them. What is the primary role of Kafka Exporter in this monitoring setup?

Great job! You've learned how to integrate Prometheus and Grafana into your monitoring strategy for Spring Boot Kafka applications and Kafka itself.

• Prometheus scrapes metrics from endpoints.
• Micrometer/Actuator expose Spring Boot metrics.
• Kafka Exporter makes Kafka broker metrics available.
• Grafana visualizes these collected metrics.

This powerful combination provides deep insights into your system's health and performance!