Skip to main content

Monitor Java applications

Choose Actuator, Micrometer, JMX Exporter or built-in application metrics, then scrape them with Categraf or Prometheus.

Last reviewed

By the end of this page, you will choose a suitable Java metrics endpoint and connect JVM or application metrics to Nightingale. Nightingale v9 does not scrape Java processes itself. Use Categraf v0.5.17, Prometheus or vmagent.

Actuator​

For a Spring Boot application, prefer Actuator. This example uses dependency management from Spring Boot 3.5.x:

<dependency>
  <groupId>org.springframework.boot</groupId>
  <artifactId>spring-boot-starter-actuator</artifactId>
</dependency>
<dependency>
  <groupId>io.micrometer</groupId>
  <artifactId>micrometer-registry-prometheus</artifactId>
</dependency>

Let the application's Spring Boot BOM manage versions instead of mixing Micrometer release lines. Expose the endpoint in application.properties:

management.server.port=8081
management.endpoints.web.exposure.include=health,prometheus

Allow the collector to access the management port. Adjust paths and credentials for Spring Security, proxies and custom application contexts.

curl --fail http://127.0.0.1:8081/actuator/prometheus

Expected result: Prometheus text with JVM metrics, not a login page, JSON error or 404. See the Spring Boot 3.5 metrics documentation.

Micrometer​

Without Spring Boot, use micrometer-registry-prometheus, register application meters with a shared PrometheusMeterRegistry, and serve its scrape() output over HTTP.

Package names vary between Micrometer and Prometheus client versions. Follow the official example for your dependencies instead of copying old imports.

A registry without an HTTP endpoint cannot be scraped. Adding a registry dependency without binding JVM meters does not create all JVM metrics automatically.

JMX​

For an application you cannot modify but that exposes MBeans, use the JMX Exporter Java Agent. This HTTP-mode example uses JMX Exporter 1.3.0. Put the jar and jmx.yaml in the working directory:

rules:
  - pattern: ".*"
java -javaagent:./jmx_prometheus_javaagent-1.3.0.jar=9404:./jmx.yaml -jar ./app.jar
curl --fail http://127.0.0.1:9404/metrics

Expected result: port 9404 returns Prometheus metrics. The broad rule is for initial verification; narrow it to required MBeans and restrict endpoint access for production. Use the 1.3.0 documentation for the package and options.

Built-in component metrics​

Reuse a component's existing Prometheus endpoint rather than adding a second JMX collection path. Some middleware and service frameworks already expose /metrics; verify the format and authentication first. Metric names and labels vary, so choose matching dashboards and rules.

Connect and verify​

Add the actual endpoint to conf/input.prometheus/prometheus.toml. This example matches the Actuator configuration above:

[[instances]]
urls = ["http://127.0.0.1:8081/actuator/prometheus"]
url_label_key = "instance"
url_label_value = "{{.Host}}"
./categraf --test --inputs prometheus

Expected result: metrics such as jvm_ without scraping or parsing errors. Restart collection and check that new samples arrive in Nightingale's destination store. If Prometheus or vmagent already scrapes the application, register its store as a datasource without adding another scrape path.

Next​

On this page