Monitoring .NET Applications with Prometheus and OpenTelemetry

Monitoring C# (.NET) Applications: From Metrics to Alerting

Modern .NET applications require constant monitoring of performance, availability, and the state of internal processes. In this article, we will explore key tools for organizing monitoring: Prometheus for collecting metrics, OpenTelemetry for tracing, and Serilog for structured logging. You will learn how to implement observability in your C# application.

Why Monitor .NET Applications?

Lack of monitoring turns debugging into guesswork. Without metrics, you won't know about increased response times; without traces, you won't find the bottleneck in the call chain; without logs, you won't understand the cause of an error. Properly configured monitoring allows you to:

• Detect anomalies before they affect users.
• Analyze the performance of individual components (databases, APIs, cache).
• Collect business metrics (number of orders, active sessions).
• Automate incident alerts.

Tools for Monitoring .NET

We will use three main components:

• Prometheus — a system for collecting and storing metrics with a powerful PromQL query language.
• OpenTelemetry — a standard for collecting telemetry (traces, metrics, logs).
• Serilog — a structured logging library for .NET.

Step 1: Installing and Configuring Prometheus with .NET

To integrate Prometheus with C#, use the prometheus-net package. Install it via NuGet:

dotnet add package prometheus-net

Then add middleware to your Program.cs:

using Prometheus;

var builder = WebApplication.CreateBuilder(args);var app = builder.Build();

// Enable the /metrics endpoint for Prometheusapp.UseHttpMetrics();app.MapMetrics();

app.MapGet("/", () => "Hello World!");app.Run();

After starting the application, navigate to http://localhost:5000/metrics. You will see standard metrics: http_requests_total, http_request_duration_seconds, and others.

Step 2: Creating Custom Metrics

Often you need to track business indicators. For example, the number of processed orders:

using Prometheus;

public class OrderMetrics{    private static readonly Counter OrdersCounter = Metrics        .CreateCounter("orders_processed_total", "Total number of processed orders");

    public static void RecordOrder()    {        OrdersCounter.Inc();    }}

// Usage in a controller[ApiController][Route("api/orders")]public class OrdersController : ControllerBase{    [HttpPost]    public IActionResult CreateOrder()    {        // ... order creation logic        OrderMetrics.RecordOrder();        return Ok();    }}

Now the /metrics endpoint will show the line orders_processed_total 1.

Step 3: Tracing with OpenTelemetry

Tracing helps understand how much time each stage of request processing takes. Install the packages:

dotnet add package OpenTelemetry.Extensions.Hostingdotnet add package OpenTelemetry.Instrumentation.AspNetCoredotnet add package OpenTelemetry.Exporter.Console

Configure OpenTelemetry in Program.cs:

using OpenTelemetry.Trace;

var builder = WebApplication.CreateBuilder(args);

builder.Services.AddOpenTelemetry()    .WithTracing(tracing => tracing        .AddAspNetCoreInstrumentation()        .AddConsoleExporter());

var app = builder.Build();app.Run();

Now every HTTP request will automatically create a trace with execution time information. For export