Docs » Instrument serverless functions for Splunk Observability Cloud » Instrument AWS Lambda functions for Splunk Observability Cloud » Instrument your AWS Lambda function for Splunk Observability Cloud » Instrument your .NET AWS Lambda function for Splunk Observability Cloud

Instrument your .NET AWS Lambda function for Splunk Observability Cloud 🔗

You can instrument a .NET AWS Lambda function to send traces to Splunk Observability Cloud using the following OpenTelemetry template. The template uses these packages:

To instrument a .NET function in AWS Lambda for Splunk APM, follow these steps:

  1. Integrate your existing AWS Lambda with the following template or start a new function using the template:

using Amazon.Lambda.Core;
using OpenTelemetry;
using OpenTelemetry.Exporter;
using OpenTelemetry.Instrumentation.AWSLambda;
using OpenTelemetry.Resources;
using OpenTelemetry.Trace;
using System.Diagnostics;

namespace DotNetInstrumentedLambdaExample;

public class Function
{
   public static readonly TracerProvider TracerProvider;

   static Function()
   {
      TracerProvider = ConfigureSplunkTelemetry()!;
   }

   // Note: Do not forget to point function handler to here.
   public string TracingFunctionHandler(string input, ILambdaContext context)
      => AWSLambdaWrapper.Trace(TracerProvider, FunctionHandler, input, context);

   public string FunctionHandler(string input, ILambdaContext context)
   {
      // TODO: Your function handler code here
   }

   private static TracerProvider ConfigureSplunkTelemetry()
   {
      var serviceName = Environment.GetEnvironmentVariable("AWS_LAMBDA_FUNCTION_NAME") ?? "Unknown";
      var accessToken = Environment.GetEnvironmentVariable("SPLUNK_ACCESS_TOKEN")?.Trim();
      var realm = Environment.GetEnvironmentVariable("SPLUNK_REALM")?.Trim();

      ArgumentNullException.ThrowIfNull(accessToken, "SPLUNK_ACCESS_TOKEN");
      ArgumentNullException.ThrowIfNull(realm, "SPLUNK_REALM");

      var builder = Sdk.CreateTracerProviderBuilder()
            // Use Add[instrumentation-name]Instrumentation to instrument missing services
            // Use Nuget to find different instrumentation libraries
            .AddHttpClientInstrumentation()
            .AddAWSInstrumentation()
            // Use AddSource to add your custom DiagnosticSource source names
            //.AddSource("My.Source.Name")
            .SetSampler(new AlwaysOnSampler())
            .AddAWSLambdaConfigurations(opts => opts.DisableAwsXRayContextExtraction = true)
            .ConfigureResource(configure => configure
                  .AddService(serviceName, serviceVersion: "1.0.0")
                  // Different resource detectors can be found at
                  // https://github.com/open-telemetry/opentelemetry-dotnet-contrib/tree/main/src/OpenTelemetry.Resources.AWS#usage
                  .AddAWSEBSDetector())
            .AddOtlpExporter(opts =>
            {
               opts.Endpoint = new Uri($"https://ingest.{realm}.signalfx.com/v2/trace/otlp");
               opts.Protocol = OtlpExportProtocol.HttpProtobuf;
               opts.Headers = $"X-SF-TOKEN={accessToken}";
            });

      return builder.Build()!;
   }
}

  1. Make sure that the main entry point is set to TracingFunctionHandler by updating the function-handler field in the aws-lambda-tools-defaults.json file to <project-name>::<class-namespace-with-class-name>::TracingFunctionHandler. You can also do this using the AWS web console, by changing the handler in Runtime settings.

    The following is an example of a aws-lambda-tools-defaults.json file with the function handler set to TracingFunctionHandler. Don’t paste the contents of the example into your file, as most of it won’t match your environment. The part that must match is TracingFunctionHandler.

    {
    "Information": [
        "This file provides default values for the deployment wizard inside Visual Studio and the AWS Lambda commands added to the .NET Core CLI.",
        "To learn more about the Lambda commands with the .NET Core CLI execute the following command at the command line in the project root directory.",
        "dotnet lambda help",
        "All the command line options for the Lambda command can be specified in this file."
    ],
    "profile": "default",
    "region": "us-west-2",
    "configuration": "Release",
    "function-architecture": "x86_64",
    "function-runtime": "dotnet8",
    "function-memory-size": 512,
    "function-timeout": 30,
    "function-handler": "AWSLambdaSample::AWSLambdaSample.Function::TracingFunctionHandler"
}

  2. The template expects the following environment variables:

    • AWS_LAMBDA_FUNCTION_NAME: Name of your AWS Lambda function

    • SPLUNK_ACCESS_TOKEN: Your Splunk ingest access token

    • SPLUNK_REALM: Your Splunk ingest realm, for example us0

  3. The template also contains the following customization points in ConfigureSplunkTelemetry():

    • Add a custom instrumentation library to support other third-party libraries. You can search for libraries using NuGet and strings starting with OpenTelemetry.Instrumentation..

    • Some libraries already have the System.Diagnostics.DiagnosticSource module built in. Use the .AddSource() method to include a custom DiagnosticSource name.

    • The AWS package contains multiple ResourceDetectors elements that help describe your environment. Select a detector for your use case.

  4. Add your code to the FunctionHandler function as the default AWS template expects.

This page was last updated on Oct 01, 2024.