Métricas do Kafka (JMX) e verificações de integridade
Entenda as principais métricas de intermediários e clientes Kafka expostas por JMX e saiba como realizar verificações de integridade nas suas aplicações Spring Boot.
Métricas do Kafka (JMX) e verificações de integridade é uma aula grátis de Advanced Spring Boot 4: Event-Driven Architecture (Kafka) no CoddyKit. Esta é a aula 1 de 4. Você pode ler a aula completa abaixo gratuitamente — depois pratica ao vivo no navegador com um editor de código integrado e um tutor de IA 24/7. Faz parte do caminho de aprendizado de Advanced Spring Boot 4: Event-Driven Architecture (Kafka), e seu progresso é sincronizado entre a web e o app CoddyKit. O curso de Advanced Spring Boot 4: Event-Driven Architecture (Kafka) inclui 4 aulas no total.
Partes desta aula ainda não foram traduzidas e aparecem em inglês.
Why Monitor Kafka & Apps?
In event-driven systems with Kafka, understanding the health and performance of your brokers and applications is crucial. Monitoring helps you detect issues early, optimize resource usage, and ensure reliable message processing.
Without proper monitoring, you'd be flying blind, unaware of potential bottlenecks, outages, or data loss risks. It's like driving a car without a dashboard!
Introducing JMX for Java Apps
JMX stands for Java Management Extensions. It's a standard technology for monitoring and managing Java applications. Kafka, being a Java application, exposes a wealth of operational data through JMX.
JMX uses objects called MBeans (Managed Beans) to expose attributes (data) and operations (actions) of an application. These MBeans provide insights into everything from memory usage to Kafka-specific metrics.
Key Kafka Broker JMX Metrics
Kafka brokers expose numerous JMX metrics that are vital for monitoring. Here are a few examples:
- MessagesInPerSec: The rate of messages produced to topics on the broker.
- BytesInPerSec/BytesOutPerSec: Network throughput for incoming/outgoing data.
- RequestPerSec: Rate of produce, fetch, or other requests handled by the broker.
- ActiveControllerCount: Indicates which broker is the cluster controller (should be 1).
Monitoring these helps you understand load, network usage, and cluster stability.
Accessing JMX Metrics
You can access JMX metrics in several ways:
- JConsole/JVisualVM: GUI tools bundled with the JDK that connect to running Java processes.
- Prometheus JMX Exporter: A popular agent that scrapes JMX metrics and exposes them in a Prometheus-compatible format.
- Programmatic Access: Using Java code to connect to the MBeanServer and query MBeans directly.
For large-scale monitoring, integrating with tools like Prometheus and Grafana is common, which we'll cover later!
Spring Boot Actuator Health
For Spring Boot applications, Actuator provides production-ready features, including powerful health check endpoints. The primary endpoint is /actuator/health.
This endpoint aggregates the health status of various components within your application, including database connections, disk space, and crucially, external dependencies like Kafka.
Enabling Actuator Endpoints
To expose Actuator endpoints, you need to add the spring-boot-starter-actuator dependency and configure your application.properties:
management.endpoints.web.exposure.include=*: Exposes all Actuator endpoints over HTTP.management.endpoint.health.show-details=always: Shows full health details, not just UP/DOWN status.
This allows you to query http://localhost:8080/actuator/health (or your app's port) to see the aggregated health.
Building Custom Health Checks
While Actuator provides out-of-the-box health checks, you often need custom ones for specific application logic or unique external dependencies. For a Kafka-integrated app, a custom health check can verify active Kafka connectivity.
You can create a custom health check by implementing Spring Boot's HealthIndicator interface. This gives you precise control over what 'healthy' means for your application's Kafka integration.
Custom Kafka Health Check
This Spring Boot example demonstrates a custom HealthIndicator that checks if a KafkaTemplate bean is available, implying successful Kafka configuration and potential connectivity.
Dependencies: Add spring-boot-starter-web, spring-boot-starter-actuator, and spring-kafka to your project's dependencies.
application.properties:
spring.kafka.bootstrap-servers=localhost:9092management.endpoints.web.exposure.include=*management.endpoint.health.show-details=always
Run this application and visit http://localhost:8080/actuator/health to see its status, including the Kafka check.
package com.coddykit.demo;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.boot.actuate.health.Health;
import org.springframework.boot.actuate.health.HealthIndicator;
import org.springframework.context.annotation.Bean;
import org.springframework.kafka.core.KafkaTemplate;
import org.springframework.beans.factory.annotation.Autowired;
@SpringBootApplication
public class DemoApplication {
public static void main(String[] args) {
SpringApplication.run(DemoApplication.class, args);
}
/**
* Custom HealthIndicator to check Kafka connectivity.
* It checks if a KafkaTemplate bean could be successfully created.
* In a real application, consider using KafkaAdminClient
* for more robust connectivity checks (e.g., listing topics).
*/
@Bean
public HealthIndicator kafkaConnectivityHealthIndicator(
@Autowired(required = false) KafkaTemplate<String, String> kafkaTemplate) {
return () -> {
if (kafkaTemplate != null) {
// If KafkaTemplate is available, assume Kafka is reachable.
return Health.up()
.withDetail("service", "Kafka Broker")
.withDetail("status", "KafkaTemplate available")
.build();
} else {
// If KafkaTemplate is null, Kafka might not be configured or reachable.
return Health.down()
.withDetail("service", "Kafka Broker")
.withDetail("error", "KafkaTemplate bean not found/failed to create")
.build();
}
};
}
}Understanding Health Endpoint
When you access /actuator/health, you'll see a JSON response. The top-level status field indicates the overall health (e.g., UP or DOWN).
Beneath that, the components field provides detailed status for each configured health indicator, including built-in ones (like disk space) and your custom Kafka check. You'll see the UP or DOWN status for each component along with any custom details you added.
Check Your Knowledge
Let's test your understanding of monitoring Kafka and Spring Boot applications.
Lesson Summary & Beyond
Great job! You've learned about the importance of monitoring, how JMX provides deep insights into Kafka brokers, and how Spring Boot Actuator enables robust health checks for your applications.
Specifically, you now understand how to expose Actuator endpoints and implement custom HealthIndicators to verify connectivity to critical services like Kafka.
Next, we'll explore integrating these metrics with powerful visualization tools like Prometheus and Grafana for comprehensive dashboards!
Perguntas Frequentes
A aula “Métricas do Kafka (JMX) e verificações de integridade” é grátis?
Sim — o texto completo de “Métricas do Kafka (JMX) e verificações de integridade” é grátis para ler aqui na web. Para praticá-la interativamente (um editor de código integrado e um tutor de IA 24/7) e desbloquear o restante do curso de Advanced Spring Boot 4: Event-Driven Architecture (Kafka), atualize para CoddyKit PRO. O curso de Advanced Spring Boot 4: Event-Driven Architecture (Kafka) inclui 4 aulas no total.
O que vou aprender em “Métricas do Kafka (JMX) e verificações de integridade”?
Entenda as principais métricas de intermediários e clientes Kafka expostas por JMX e saiba como realizar verificações de integridade nas suas aplicações Spring Boot. Você pratica Advanced Spring Boot 4: Event-Driven Architecture (Kafka) com código prático que executa diretamente no navegador, e um tutor de IA 24/7 responde suas dúvidas enquanto trabalha na aula.
Preciso ter experiência prévia para começar Advanced Spring Boot 4: Event-Driven Architecture (Kafka)?
Nenhuma experiência prévia é necessária. Advanced Spring Boot 4: Event-Driven Architecture (Kafka) no CoddyKit é estruturado para alunos iniciantes até avançados, então você pode começar aqui ou desde o início e aprender no seu ritmo. Esta é a aula 1 de 4.
Quanto tempo leva a aula “Métricas do Kafka (JMX) e verificações de integridade”?
A maioria das aulas CoddyKit leva cerca de 5–10 minutos. Cada uma é compacta e interativa, então você faz progresso constante e retoma exatamente de onde parou entre web e app.
Posso escrever e executar código nesta aula de Advanced Spring Boot 4: Event-Driven Architecture (Kafka)?
Sim. Cada aula de Advanced Spring Boot 4: Event-Driven Architecture (Kafka) inclui um editor de código integrado, então você escreve e executa código real direto no navegador e recebe feedback de IA instantaneamente — nenhuma configuração local necessária.
Todas as aulas deste curso
- Métricas do Kafka (JMX) e verificações de integridade
- Integração com Prometheus e Grafana
- Rastreamento distribuído com Sleuth/Zipkin
- Monitorando o atraso dos consumidores e configurando alertas