ilya-rochev/grafana-dashboard-generator

Annotation processor which automatically generates Grafana Dashboard JSON model based on metrics declared in application

Grafana Dashboard Generator

Annotation processor which automatically generates Grafana Dashboard JSON model based on metrics declared in application.

How to build project

run command

./mvnw clean install

How to use library

1. Add next dependencies to pom.xml

<dependency>
    <groupId>io.github.ilya-rochev</groupId>
    <artifactId>grafana-dashboard-annotations-processor</artifactId>
    <version>1.0</version>
</dependency>

<dependency>
    <groupId>io.github.ilya-rochev</groupId>
    <artifactId>prometheus-metric-spring-boot-starter</artifactId>
    <version>1.0</version>
</dependency>

2. Migrate creating Counter and Timer beans to processed by prometheus-metric-spring-boot-starter annotation i.r.grafana.json.annotations.promethues.@Metric.

If you create metrics like this

@Configuration
@RequiredArgsConstructor
public class MetricConfig {

    private final MeterRegistry meterRegistry;

    @Bean
    public Counter savedObjectsCounter() {
        return meterRegistry.counter("counter_as_field",  "type", "saved");
    }
    
    ...  
}

Change creating metrics declarative way using @Metric annotation

@Service
@RequiredArgsConstructor
@ParametersAreNonnullByDefault
public class ExternalCounterServiceStub {

    @Metric(value = "counter_as_field", extraTags = {"type", "saved"}, description = "Saved object counter")
    private final Counter savedObjectsCounter;
    
    ...
}   

Notice: prometheus-metric-spring-boot-starter processes annotation @Metric during starting application. It creates Timer or Counter bean with name and tags you provided as annotation parameters.

3. Declare you Grafana dashboard using annotations from package i.r.grafana.json.annotations

@Dashboard(title = "Demo application metrics", panels = {
        @Panel(metric = "timer_as_field", title = "External service invocation time (FIELDS)"),
        @Panel(metric = "timer_as_annotation", title = "External service invocation time (ANNOTATIONS)"),
        @Panel(metric = "counter_as_field", title = "External objects counters (FIELDS)"),
        @Panel(metric = "counter_as_annotation", title = "External objects counters (ANNOTATION)")}
)
public class GrafanaDashboardConfiguration { }

Notice: grafana-dashboard-annotations-processor processes annotations from package i.r.grafana.json.annotations during compile. It generates dashboard model and injects metrics you declared using @Metric annotation;

4. Build your project. Created grafana json model will be stored in target\classes\dasboard_{uuid}.json file, for example - dashboard_221a4f4c-2d60-4ec0-9fa5-7c828fee81eb.json

5. Run your project and import generated json file to grafana. Enjoy to view you custom application metrics.

Demo application

Project provides demo application to be informed how to use library in practice

Requires:

- "docker" runned at your machine;

1. Start locally grafana and prometheus in docker:

docker-compose up -d

2. Export generated metrics to Grafana

3. Start demo application;

4. Test demo application with postman collection docs/Demo_Application_GRAFANA.postman_collection.json

5. See metrics in Grafana