Skip to content

Send your first trace

In about five minutes you will send real telemetry into IAPM and watch a system take shape in the Grid: services appearing as you feed them, connected by the calls between them. You will use a ready-made trace generator, so there is no application to write and nothing to deploy. This is the known-good baseline: one command, real OTLP traces, your data in IAPM.

You are not instrumenting your own application yet. That comes next, and it is easier once you have seen what "working" looks like.

What you will learn

  • How to point an OpenTelemetry exporter at IAPM using your grid's API key
  • What a healthy stream of traces looks like when it arrives
  • Where your services and traces show up in IAPM

Prerequisites

  • A subscription and your own grid. Sending your own telemetry requires an active subscription (the free Start plan works) with a provisioned grid. Provisioning a grid requires a valid credit card, even on Start. If you have not done this yet, complete Getting Started first. (Just want to look around? The demo grids need no subscription.)
  • The grid's API key. You will copy this in Step 1.
  • Go 1.21 or later installed, so you can run the generator with a single command. (No Go knowledge is needed; you are only running a tool written in it.)

Step 1: Get your grid API key

Your API key authenticates your telemetry with IAPM. Each grid has its own key.

  1. Sign in at portal.iapm.app (the sign-in chooser). After you choose a method you land in the web app: my.iapm.app if you signed in with Email or GitHub, azure.iapm.app if you signed in with Microsoft. Go to Administration → Grids.
  2. Click the Instrument button on the grid you want to send traces to.
  3. Copy the API key from the wizard.

Keep the key somewhere safe for the next step. Treat it like a password: do not commit it to source control.

For more detail, see the API Keys reference.

Step 2: Install the trace generator

The OpenTelemetry Trace Generator is a single binary that simulates a realistic, multi-service application and sends its traces over OTLP. Install it with:

go install github.com/ImmersiveFusion/if-opentelemetry-tracegen/cmd/tracegen@latest

This puts a tracegen command on your system.

What you should see: the command completes with no output. Run tracegen -help to confirm it is installed; you will see the list of available flags.

Step 3: Send traces to IAPM

Now point the generator at IAPM. Replace YOUR_IAPM_KEY with the key you copied in Step 1:

tracegen -endpoint otlp.iapm.app:443 -headers "api-key=YOUR_IAPM_KEY" -complexity light

A note on each part:

  • -endpoint otlp.iapm.app:443 sends to IAPM's OTLP endpoint over gRPC.
  • -headers "api-key=YOUR_IAPM_KEY" authenticates with your grid key. The header name is api-key, lowercase, exactly.
  • -complexity light generates a clean, ten-service topology: enough to be interesting, small enough to read at a glance. (You can rerun later with normal or heavy for a larger graph.)

What you should see: the generator starts emitting traces and logs to your terminal as it simulates an e-commerce platform: a web frontend, an API gateway, an order service, a payment service, and a handful of others, exercising flows like Create Order and User Login.

Sending to a local collector first?

To try the generator against a local Jaeger, Tempo, or Collector before involving IAPM, drop the headers and use tracegen -insecure (defaults to localhost:4317). Then come back and run the IAPM command above.

Step 4: See your trace in IAPM

Switch to IAPM and open the grid you used in Step 1.

What you should see: within a couple of minutes, services begin appearing in the Grid: nodes for web-frontend, api-gateway, order-service, and the rest, connected by the calls between them. Open a service and drill into a trace to see its span waterfall: the ordered timeline of every operation in a single request.

You have now sent your first trace into IAPM and watched it arrive. That is the whole loop: telemetry leaves your machine, IAPM receives it, and the Grid shows you the shape of the system.

What you did

In about five minutes you:

  • Authenticated a telemetry source to IAPM with a grid API key
  • Sent real OTLP traces using a ready-made generator
  • Watched services and traces appear in the Grid

Next steps

  • Instrument your own application. Swap the generator for your real service: see Instrument Your Application for language-specific guides (.NET, Java, Python, Go, Node.js).
  • Explore failures. Rerun the generator with -errors 5 to inject failures and watch how they propagate through the Grid.
  • Ask Tessa about it. With data flowing, open the AI Assistant and ask "How is my system doing right now?" to see IAPM's diagnostics in action. Tessa is in ALPHA.