Skip to content

Latest commit

 

History

History
417 lines (288 loc) · 25.1 KB

opentelemetry-enable.md

File metadata and controls

417 lines (288 loc) · 25.1 KB
title description ms.topic ms.date ms.devlang ms.custom ms.reviewer
Enable OpenTelemetry in Application Insights
Learn how to enable OpenTelemetry (OTel) data collection in Application Insights for .NET, Java, Node.js, and Python applications using the Azure Monitor OpenTelemetry Distro.
conceptual
02/28/2025
csharp
devx-track-dotnet, devx-track-extended-java, devx-track-python
mmcc

Enable Azure Monitor OpenTelemetry for .NET, Node.js, Python, and Java applications

This article describes how to enable and configure OpenTelemetry-based data collection within Azure Monitor Application Insights. The Azure Monitor OpenTelemetry Distro:

  • Provides an OpenTelemetry distribution which includes support for features specific to Azure Monitor.
  • Enables automatic telemetry by including OpenTelemetry instrumentation libraries for collecting traces, metrics, logs, and exceptions.
  • Allows collecting custom telemetry.
  • Supports Live Metrics to monitor and collect more telemetry from live, in-production web applications.

For more information about the advantages of using the Azure Monitor OpenTelemetry Distro, see Why should I use the Azure Monitor OpenTelemetry Distro.

To learn more about collecting data using OpenTelemetry, check out Data Collection Basics or the OpenTelemetry FAQ.

OpenTelemetry release status

OpenTelemetry offerings are available for .NET, Node.js, Python, and Java applications. For a feature-by-feature release status, see the FAQ.

Enable OpenTelemetry with Application Insights

Follow the steps in this section to instrument your application with OpenTelemetry. Select a tab for langauge-specific instructions.

Note

.NET covers multiple scenarios, including classic ASP.NET, console apps, Windows Forms (WinForms), and more.

Prerequisites

[!div class="checklist"]

ASP.NET Core

[!div class="checklist"]

Tip

If you're migrating from the Application Insights Classic API, see our migration documentation.

.NET

[!div class="checklist"]

Tip

If you're migrating from the Application Insights Classic API, see our migration documentation.

Java

[!div class="checklist"]

  • A Java application using Java 8+

Java native

[!div class="checklist"]

  • A Java application using GraalVM 17+

Node.js

[!div class="checklist"]

Note

If you don't rely on any properties listed in the not-supported table, the ApplicationInsights shim will be your easiest path forward once out of beta.

If you rely on any of those properties, proceed with the Azure Monitor OpenTelemetry Distro. We'll provide a migration guide soon.

Tip

If you're migrating from the Application Insights Classic API, see our migration documentation.

Python

[!div class="checklist"]

  • Python Application using Python 3.8+

Tip

If you're migrating from OpenCensus, see our migration documentation.

Install the client library

ASP.NET Core

Install the latest Azure.Monitor.OpenTelemetry.AspNetCore NuGet package:

dotnet add package Azure.Monitor.OpenTelemetry.AspNetCore

.NET

Install the latest Azure.Monitor.OpenTelemetry.Exporter NuGet package:

dotnet add package Azure.Monitor.OpenTelemetry.Exporter

Java

Download the latest applicationinsights-agent-3.7.1.jar file.

Warning

If you are upgrading from an earlier 3.x version, you may be impacted by changing defaults or slight differences in the data we collect. For more information, see the migration section in the release notes. 3.5.0, 3.4.0, 3.3.0, 3.2.0, and 3.1.0

Java native

For Spring Boot native applications:

For Quarkus native applications:

Node.js

Install the latest @azure/monitor-opentelemetry package:

npm install @azure/monitor-opentelemetry

The following packages are also used for some specific scenarios described later in this article:

npm install @opentelemetry/api
npm install @opentelemetry/sdk-metrics
npm install @opentelemetry/resources
npm install @opentelemetry/semantic-conventions
npm install @opentelemetry/sdk-trace-base

Python

Install the latest azure-monitor-opentelemetry PyPI package:

pip install azure-monitor-opentelemetry

Modify your application

ASP.NET Core

Import the Azure.Monitor.OpenTelemetry.AspNetCore namespace, add OpenTelemetry, and configure it to use Azure Monitor in your program.cs class:

// Import the Azure.Monitor.OpenTelemetry.AspNetCore namespace.
using Azure.Monitor.OpenTelemetry.AspNetCore;

var builder = WebApplication.CreateBuilder(args);

// Add OpenTelemetry and configure it to use Azure Monitor.
builder.Services.AddOpenTelemetry().UseAzureMonitor();

var app = builder.Build();

app.Run();

.NET

Add the Azure Monitor Exporter to each OpenTelemetry signal in the program.cs class:

// Create a new tracer provider builder and add an Azure Monitor trace exporter to the tracer provider builder.
// It is important to keep the TracerProvider instance active throughout the process lifetime.
// See https://github.com/open-telemetry/opentelemetry-dotnet/tree/main/docs/trace#tracerprovider-management
var tracerProvider = Sdk.CreateTracerProviderBuilder()
    .AddAzureMonitorTraceExporter();

// Add an Azure Monitor metric exporter to the metrics provider builder.
// It is important to keep the MetricsProvider instance active throughout the process lifetime.
// See https://github.com/open-telemetry/opentelemetry-dotnet/tree/main/docs/metrics#meterprovider-management
var metricsProvider = Sdk.CreateMeterProviderBuilder()
    .AddAzureMonitorMetricExporter();

// Create a new logger factory.
// It is important to keep the LoggerFactory instance active throughout the process lifetime.
// See https://github.com/open-telemetry/opentelemetry-dotnet/tree/main/docs/logs#logger-management
var loggerFactory = LoggerFactory.Create(builder =>
{
    builder.AddOpenTelemetry(logging =>
    {
        logging.AddAzureMonitorLogExporter();
    });
});

Note

For more information, see the getting-started tutorial for OpenTelemetry .NET

Java

Autoinstrumentation is enabled through configuration changes. No code changes are required.

Point the Java virtual machine (JVM) to the jar file by adding -javaagent:"path/to/applicationinsights-agent-3.7.1.jar" to your application's JVM args.

Note

Sampling is enabled by default at a rate of 5 requests per second, aiding in cost management. Telemetry data may be missing in scenarios exceeding this rate. For more information on modifying sampling configuration, see sampling overrides.

Tip

For scenario-specific guidance, see Get Started (Supplemental).

Tip

If you develop a Spring Boot application, you can optionally replace the JVM argument by a programmatic configuration. For more information, see Using Azure Monitor Application Insights with Spring Boot.

Java native

Autoinstrumentation is enabled through configuration changes. No code changes are required.

Node.js

// Import the `useAzureMonitor()` function from the `@azure/monitor-opentelemetry` package.
const { useAzureMonitor } = require("@azure/monitor-opentelemetry");

// Call the `useAzureMonitor()` function to configure OpenTelemetry to use Azure Monitor.
useAzureMonitor();

Python

import logging
# Import the `configure_azure_monitor()` function from the
# `azure.monitor.opentelemetry` package.
from azure.monitor.opentelemetry import configure_azure_monitor

# Configure OpenTelemetry to use Azure Monitor with the 
# APPLICATIONINSIGHTS_CONNECTION_STRING environment variable.
configure_azure_monitor(
    logger_name="<your_logger_namespace>",  # Set the namespace for the logger in which you would like to collect telemetry for if you are collecting logging telemetry. This is imperative so you do not collect logging telemetry from the SDK itself.
)
logger = logging.getLogger("<your_logger_namespace>")  # Logging telemetry will be collected from logging calls made with this logger and all of it's children loggers.

Copy the connection string from your Application Insights resource

The connection string is unique and specifies where the Azure Monitor OpenTelemetry Distro sends the telemetry it collects.

Tip

If you don't already have an Application Insights resource, create one following this guide. We recommend you create a new resource rather than using an existing one.

To copy the connection string:

  1. Go to the Overview pane of your Application Insights resource.
  2. Find your connection string.
  3. Hover over the connection string and select the Copy to clipboard icon.

:::image type="content" source="media/migrate-from-instrumentation-keys-to-connection-strings/migrate-from-instrumentation-keys-to-connection-strings.png" alt-text="Screenshot that shows Application Insights overview and connection string." lightbox="media/migrate-from-instrumentation-keys-to-connection-strings/migrate-from-instrumentation-keys-to-connection-strings.png":::

Paste the connection string in your environment

To paste your connection string, select from the following options:

Important

We recommend setting the connection string through code only in local development and test environments.

For production, use an environment variable or configuration file (Java only).

  • Set via environment variable - recommended

    Replace <Your connection string> in the following command with your connection string.

    APPLICATIONINSIGHTS_CONNECTION_STRING=<Your connection string>

  • Set via configuration file - Java only

    Create a configuration file named applicationinsights.json, and place it in the same directory as applicationinsights-agent-3.7.1.jar with the following content:

    {
  "connectionString": "<Your connection string>"
}

    Replace <Your connection string> in the preceding JSON with your unique connection string.

  • Set via code - ASP.NET Core, Node.js, and Python only

    See connection string configuration for an example of setting connection string via code.

Note

If you set the connection string in multiple places, the environment variable will be prioritized in the following order:

  1. Code
  2. Environment variable
  3. Configuration file

Confirm data is flowing

Run your application, then open Application Insights in the Azure portal. It might take a few minutes for data to show up.

:::image type="content" source="media/opentelemetry/server-requests.png" alt-text="Screenshot of the Application Insights Overview tab with server requests and server response time highlighted.":::

Application Insights is now enabled for your application. The following steps are optional and allow for further customization.

Important

If you have two or more services that emit telemetry to the same Application Insights resource, you're required to set Cloud Role Names to represent them properly on the Application Map.

As part of using Application Insights instrumentation, we collect and send diagnostic data to Microsoft. This data helps us run and improve Application Insights. To learn more, see Statsbeat in Azure Application Insights.

Sample applications

Azure Monitor OpenTelemetry sample applications are available for all supported languages:

Next steps

ASP.NET Core

.NET

Java

Java native

Node.js

Python