Exporters

Send telemetry to the OpenTelemetry Collector to make sure it’s exported correctly. Using the Collector in production environments is a best practice. To visualize your telemetry, export it to a backend such as Jaeger, Zipkin, Prometheus, or a vendor-specific backend.

Available exporters

The registry contains a list of exporters for Java.

Among exporters, OpenTelemetry Protocol (OTLP) exporters are designed with the OpenTelemetry data model in mind, emitting OTel data without any loss of information. Furthermore, many tools that operate on telemetry data support OTLP (such as Prometheus, Jaeger, and most vendors), providing you with a high degree of flexibility when you need it. To learn more about OTLP, see OTLP Specification.

This page covers the main OpenTelemetry Java exporters and how to set them up.

OTLP

Collector Setup

To try out and verify your OTLP exporters, you can run the collector in a docker container that writes telemetry directly to the console.

In an empty directory, create a file called collector-config.yaml with the following content:

receivers:
  otlp:
    protocols:
      grpc:
        endpoint: 0.0.0.0:4317
      http:
        endpoint: 0.0.0.0:4318
exporters:
  debug:
    verbosity: detailed
service:
  pipelines:
    traces:
      receivers: [otlp]
      exporters: [debug]
    metrics:
      receivers: [otlp]
      exporters: [debug]
    logs:
      receivers: [otlp]
      exporters: [debug]

Now run the collector in a docker container:

docker run -p 4317:4317 -p 4318:4318 --rm -v $(pwd)/collector-config.yaml:/etc/otelcol/config.yaml otel/opentelemetry-collector

This collector is now able to accept telemetry via OTLP. Later you may want to configure the collector to send your telemetry to your observability backend.

Dependencies

If you want to send telemetry data to an OTLP endpoint (like the OpenTelemetry Collector, Jaeger or Prometheus), there are multiple OTLP options available, each catering to different use cases. For most users, the default artifact will suffice and be the most simple:

dependencies {
  implementation("io.opentelemetry:opentelemetry-exporter-otlp:1.42.0")
  implementation("io.opentelemetry:opentelemetry-sdk:1.42.0")
  implementation("io.opentelemetry.semconv:opentelemetry-semconv:1.27.0-alpha")
}

<project>
    <dependencies>
        <dependency>
            <groupId>io.opentelemetry</groupId>
            <artifactId>opentelemetry-exporter-otlp</artifactId>
        </dependency>
        <dependency>
            <groupId>io.opentelemetry</groupId>
            <artifactId>opentelemetry-sdk</artifactId>
        </dependency>
        <dependency>
            <groupId>io.opentelemetry.semconv</groupId>
            <artifactId>opentelemetry-semconv</artifactId>
            <version>1.27.0-alpha</version>
        </dependency>
    </dependencies>
</project>

Under the hood, there are two protocol options supported, each with different “sender” implementations.

  • grpc - gRPC implementation of OTLP exporters, represented by OtlpGrpcSpanExporter, OtlpGrpcMetricExporter, OtlpGrpcLogRecordExporter.
  • http/protobuf - HTTP with protobuf encoded payload implementation of OTLP exporters, represented by OtlpHttpSpanExporter, OtlpHttpMetricExporter, OtlpHttpLogRecordExporter.

A sender is an abstraction which allows different gRPC / HTTP client implementations to fulfill the OTLP contract. Regardless of the sender implementation, the same exporter classes are used. A sender implementation is automatically used when it is detected on the classpath. The sender implementations are described in detail below:

  • {groupId}:{artifactId} - Sender description.
  • io.opentelemetry:opentelemetry-exporter-sender-okhttp - The default sender, included automatically with opentelemetry-exporter-otlp and bundled with the OpenTelemetry Java agent. This includes an OkHttp based implementation for both the grpc and http/protobuf versions of the protocol, and will be suitable for most users. However, OkHttp has a transitive dependency on kotlin which is problematic in some environments.
  • io.opentelemetry:opentelemetry-exporter-sender-jdk - This sender includes a JDK 11+ HttpClient based implementation for the http/protobuf version of the protocol. It requires zero additional dependencies, but requires Java 11+. To use, include the artifact and explicitly exclude the default io.opentelemetry:opentelemetry-exporter-sender-okhttp dependency.
  • io.opentelemetry:opentelemetry-exporter-sender-grpc-managed-channel - This sender includes a grpc-java based implementation for the grpc version of the protocol. To use, include the artifact, explicitly exclude the default io.opentelemetry:opentelemetry-exporter-sender-okhttp dependency, and include one of the gRPC transport implementations.

Usage

Next, configure the exporter to point at an OTLP endpoint.

If you use SDK autoconfiguration all you need to do is update your environment variables:

env OTEL_EXPORTER_OTLP_ENDPOINT=http://example:4317 java -jar ./build/libs/java-simple.jar

Note, that in the case of exporting via OTLP you do not need to set OTEL_TRACES_EXPORTER, OTEL_METRICS_EXPORTER and OTEL_LOGS_EXPORTER since otlp is their default value

In the case of [manual configuration] you can update the example app like the following:

package otel;

import io.opentelemetry.api.OpenTelemetry;
import io.opentelemetry.api.trace.propagation.W3CTraceContextPropagator;
import io.opentelemetry.context.propagation.ContextPropagators;
import io.opentelemetry.exporter.otlp.logs.OtlpGrpcLogRecordExporter;
import io.opentelemetry.exporter.otlp.metrics.OtlpGrpcMetricExporter;
import io.opentelemetry.exporter.otlp.trace.OtlpGrpcSpanExporter;
import io.opentelemetry.sdk.OpenTelemetrySdk;
import io.opentelemetry.sdk.logs.SdkLoggerProvider;
import io.opentelemetry.sdk.logs.export.BatchLogRecordProcessor;
import io.opentelemetry.sdk.metrics.SdkMeterProvider;
import io.opentelemetry.sdk.metrics.export.PeriodicMetricReader;
import io.opentelemetry.sdk.resources.Resource;
import io.opentelemetry.sdk.trace.SdkTracerProvider;
import io.opentelemetry.sdk.trace.export.BatchSpanProcessor;
import io.opentelemetry.semconv.ServiceAttributes;
import org.springframework.boot.Banner;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.context.annotation.Bean;

@SpringBootApplication
public class DiceApplication {
  public static void main(String[] args) {
    SpringApplication app = new SpringApplication(DiceApplication.class);
    app.setBannerMode(Banner.Mode.OFF);
    app.run(args);
  }

  @Bean
  public OpenTelemetry openTelemetry() {
    Resource resource =
        Resource.getDefault().toBuilder()
            .put(ServiceAttributes.SERVICE_NAME, "dice-server")
            .put(ServiceAttributes.SERVICE_VERSION, "0.1.0")
            .build();

    SdkTracerProvider sdkTracerProvider =
        SdkTracerProvider.builder()
            .addSpanProcessor(
                BatchSpanProcessor.builder(OtlpGrpcSpanExporter.builder().build()).build())
            .setResource(resource)
            .build();

    SdkMeterProvider sdkMeterProvider =
        SdkMeterProvider.builder()
            .registerMetricReader(
                PeriodicMetricReader.builder(OtlpGrpcMetricExporter.builder().build()).build())
            .setResource(resource)
            .build();

    SdkLoggerProvider sdkLoggerProvider =
        SdkLoggerProvider.builder()
            .addLogRecordProcessor(
                BatchLogRecordProcessor.builder(OtlpGrpcLogRecordExporter.builder().build())
                    .build())
            .setResource(resource)
            .build();

    OpenTelemetry openTelemetry =
        OpenTelemetrySdk.builder()
            .setTracerProvider(sdkTracerProvider)
            .setMeterProvider(sdkMeterProvider)
            .setLoggerProvider(sdkLoggerProvider)
            .setPropagators(ContextPropagators.create(W3CTraceContextPropagator.getInstance()))
            .buildAndRegisterGlobal();

    return openTelemetry;
  }
}

Console

To debug your instrumentation or see the values locally in development, you can use exporters writing telemetry data to the console (stdout).

If you followed the Getting Started or Manual Instrumentation guides, you already have the console exporter installed.

The LoggingSpanExporter, the LoggingMetricExporter and the SystemOutLogRecordExporter are included in the opentelemetry-exporter-logging artifact.

If you use SDK autoconfiguration all you need to do is update your environment variables:

env OTEL_TRACES_EXPORTER=logging OTEL_METRICS_EXPORTER=logging OTEL_LOGS_EXPORTER=logging  java -jar ./build/libs/java-simple.jar

Jaeger

Backend Setup

Jaeger natively supports OTLP to receive trace data. You can run Jaeger in a docker container with the UI accessible on port 16686 and OTLP enabled on ports 4317 and 4318:

docker run --rm \
  -e COLLECTOR_ZIPKIN_HOST_PORT=:9411 \
  -p 16686:16686 \
  -p 4317:4317 \
  -p 4318:4318 \
  -p 9411:9411 \
  jaegertracing/all-in-one:latest

Usage

Now following the instruction to setup the OTLP exporters.

Prometheus

To send your metric data to Prometheus, you can either enable Prometheus’ OTLP Receiver and use the OTLP exporter or you can use the Prometheus exporter, a MetricReader that starts an HTTP server that collects metrics and serialize to Prometheus text format on request.

Backend Setup

You can run Prometheus in a docker container, accessible on port 9090 by following these instructions:

Create a file called prometheus.yml with the following content:

scrape_configs:
  - job_name: dice-service
    scrape_interval: 5s
    static_configs:
      - targets: [host.docker.internal:9464]

Run Prometheus in a docker container with the UI accessible on port 9090:

docker run --rm -v ${PWD}/prometheus.yml:/prometheus/prometheus.yml -p 9090:9090 prom/prometheus --enable-feature=otlp-write-receive

Dependencies

Install the opentelemetry-exporter-prometheus artifact as a dependency for your application:

dependencies {
    implementation 'io.opentelemetry:opentelemetry-exporter-prometheus:1.42.0-alpha'
}

<project>
    <dependencies>
        <dependency>
            <groupId>io.opentelemetry</groupId>
            <artifactId>opentelemetry-exporter-prometheus</artifactId>
        </dependency>
    </dependencies>
</project>

Update your OpenTelemetry configuration to use the exporter and to send data to your Prometheus backend:

package otel;

import io.opentelemetry.exporter.prometheus.PrometheusHttpServer;
import io.opentelemetry.sdk.metrics.SdkMeterProvider;
import io.opentelemetry.sdk.resources.Resource;

public class PrometheusExporter {
  public static SdkMeterProvider create(Resource resource) {
    int prometheusPort = 9464;
    SdkMeterProvider sdkMeterProvider =
        SdkMeterProvider.builder()
            .registerMetricReader(PrometheusHttpServer.builder().setPort(prometheusPort).build())
            .setResource(resource)
            .build();

    return sdkMeterProvider;
  }
}

With the above you can access your metrics at http://localhost:9464/metrics. Prometheus or an OpenTelemetry Collector with the Prometheus receiver can scrape the metrics from this endpoint.

Zipkin

Backend Setup

You can run Zipkin on in a Docker container by executing the following command:

docker run --rm -d -p 9411:9411 --name zipkin openzipkin/zipkin

Dependencies

To send your trace data to Zipkin, you can use the ZipkinSpanExporter.

Install the opentelemetry-exporter-zipkin artifact as a dependency for your application:

dependencies {
    implementation 'io.opentelemetry:opentelemetry-exporter-zipkin:1.42.0-alpha'
}

<project>
    <dependencies>
        <dependency>
            <groupId>io.opentelemetry</groupId>
            <artifactId>opentelemetry-exporter-zipkin</artifactId>
        </dependency>
    </dependencies>
</project>

Update your OpenTelemetry configuration to use the exporter and to send data to your Zipkin backend:

package otel;

import io.opentelemetry.exporter.zipkin.ZipkinSpanExporter;
import io.opentelemetry.sdk.resources.Resource;
import io.opentelemetry.sdk.trace.SdkTracerProvider;
import io.opentelemetry.sdk.trace.export.BatchSpanProcessor;

public class ZipkinExporter {
  public static SdkTracerProvider create(Resource resource) {
    SdkTracerProvider sdkTracerProvider =
        SdkTracerProvider.builder()
            .addSpanProcessor(
                BatchSpanProcessor.builder(
                        ZipkinSpanExporter.builder()
                            .setEndpoint("http://localhost:9411/api/v2/spans")
                            .build())
                    .build())
            .setResource(resource)
            .build();

    return sdkTracerProvider;
  }
}

Custom exporters

Finally, you can also write your own exporter. For more information, see the SpanExporter Interface in the API documentation.

Batching span and log records

The OpenTelemetry SDK provides a set of default span and log record processors, that allow you to either emit spans one-by-on (“simple”) or batched. Using batching is recommended, but if you do not want to batch your spans or log records, you can use a simple processor instead as follows:

package otel;

import io.opentelemetry.sdk.logs.SdkLoggerProvider;
import io.opentelemetry.sdk.logs.export.BatchLogRecordProcessor;
import io.opentelemetry.sdk.logs.export.LogRecordExporter;
import io.opentelemetry.sdk.resources.Resource;
import io.opentelemetry.sdk.trace.SdkTracerProvider;
import io.opentelemetry.sdk.trace.export.BatchSpanProcessor;
import io.opentelemetry.sdk.trace.export.SpanExporter;

public class BatchExporter {
  public static void create(
      Resource resource, SpanExporter spanExporter, LogRecordExporter logExporter) {
    SdkTracerProvider sdkTracerProvider =
        SdkTracerProvider.builder()
            .addSpanProcessor(BatchSpanProcessor.builder(spanExporter).build())
            .setResource(resource)
            .build();

    SdkLoggerProvider sdkLoggerProvider =
        SdkLoggerProvider.builder()
            .addLogRecordProcessor(BatchLogRecordProcessor.builder(logExporter).build())
            .setResource(resource)
            .build();
  }
}

package otel;

import io.opentelemetry.sdk.logs.SdkLoggerProvider;
import io.opentelemetry.sdk.logs.export.LogRecordExporter;
import io.opentelemetry.sdk.logs.export.SimpleLogRecordProcessor;
import io.opentelemetry.sdk.resources.Resource;
import io.opentelemetry.sdk.trace.SdkTracerProvider;
import io.opentelemetry.sdk.trace.export.SimpleSpanProcessor;
import io.opentelemetry.sdk.trace.export.SpanExporter;

public class SimpleExporter {
  public static void create(
      Resource resource, SpanExporter spanExporter, LogRecordExporter logExporter) {
    SdkTracerProvider sdkTracerProvider =
        SdkTracerProvider.builder()
            .addSpanProcessor(SimpleSpanProcessor.builder(spanExporter).build())
            .setResource(resource)
            .build();

    SdkLoggerProvider sdkLoggerProvider =
        SdkLoggerProvider.builder()
            .addLogRecordProcessor(SimpleLogRecordProcessor.create(logExporter))
            .setResource(resource)
            .build();
  }
}

最終更新 July 24, 2024: Include Java example code excerpts (#4642) (99dfcdeb)