Application Insights

Application Insights is a feature of Azure Monitor that provides application performance management (APM) for your web applications and services. It automatically collects telemetry about requests, dependencies, exceptions, and performance, giving you deep visibility into how your applications behave in production.

---

What Is Application Insights?

Application Insights monitors your live applications and helps you:

• Detect anomalies — automatically identify unusual patterns in performance and failure rates
• Diagnose issues — trace requests end-to-end across services and dependencies
• Understand usage — analyse how users interact with your application
• Improve performance — identify slow dependencies, bottlenecks, and resource contention

Key Capabilities

Capability	Description
Request tracking	Monitor response times, throughput, and failure rates
Dependency tracking	Track calls to databases, HTTP APIs, Azure services
Exception tracking	Capture exceptions with full stack traces and context
Live Metrics	Real-time telemetry stream during active investigation
Availability tests	Synthetic tests that check your app from multiple Azure regions
Application Map	Visual topology of your application and its dependencies
Smart Detection	ML-powered anomaly detection for failure rates and performance
User analytics	User flows, sessions, funnels, and retention analysis

---

Instrumentation Methods

There are two ways to add Application Insights to your application:

Auto-Instrumentation

Azure can automatically collect telemetry without code changes for supported platforms:

Platform	Method
Azure App Service (.NET)	Enable in the Azure Portal — no code changes
Azure App Service (Java)	Enable Java agent in the Portal
Azure App Service (Node.js)	Enable in the Portal
Azure Functions	Built-in integration
Azure VMs	Install the Application Insights agent
AKS	Use the OpenTelemetry auto-instrumentation operator

SDK Instrumentation

For more control, add the Application Insights SDK to your application code:

.NET:

// Program.cs
var builder = WebApplication.CreateBuilder(args);
builder.Services.AddApplicationInsightsTelemetry();

Python:

from azure.monitor.opentelemetry import configure_azure_monitor

configure_azure_monitor(
    connection_string="InstrumentationKey=<key>;IngestionEndpoint=https://uksouth-1.in.applicationinsights.azure.com/"
)

JavaScript / Node.js:

const { useAzureMonitor } = require('@azure/monitor-opentelemetry');

useAzureMonitor({
  azureMonitorExporterOptions: {
    connectionString: process.env.APPLICATIONINSIGHTS_CONNECTION_STRING
  }
});

The SDK approach gives you the ability to add custom telemetry, custom properties, and fine-grained control over what is collected.

---

Telemetry Types

Application Insights collects several types of telemetry:

Telemetry Type	Table in Log Analytics	Description
Requests	AppRequests	Incoming HTTP requests to your app
Dependencies	AppDependencies	Outgoing calls to databases, HTTP APIs, Azure services
Exceptions	AppExceptions	Unhandled and tracked exceptions
Page views	AppPageViews	Client-side page load events
Custom events	AppEvents	Business events you define
Custom metrics	AppMetrics	Numerical measurements you define
Traces	AppTraces	Diagnostic log messages
Availability	AppAvailabilityResults	Synthetic test results

---

Application Map

The Application Map provides a visual topology of your application, showing:

• Each component (web app, API, database, external service)
• The number of requests and average response time between components
• Failure rates highlighted in red
• Click any component to drill into its telemetry

This is invaluable for understanding distributed applications and quickly identifying which component is causing issues.

---

Live Metrics

Live Metrics provides a real-time telemetry stream with zero latency:

• Incoming request rate and response times
• Dependency call rate and duration
• Exception rate
• CPU and memory usage
• Custom metrics and filters

Live Metrics is particularly useful during active incidents or immediately after a deployment to verify that everything is working correctly.

---

Availability Tests