Important: This documentation is about an older version. It's relevant only to the release noted, many of the features and functions have been updated or replaced. Please view the current version.

Documentationbreadcrumb arrow Grafana documentationbreadcrumb arrow Explorebreadcrumb arrow Simplified explorationbreadcrumb arrow Traces Drilldownbreadcrumb arrow Get startedbreadcrumb arrow Investigation walkthrough
Enterprise Open source

Investigation walkthrough

Follow this step-by-step walkthrough to investigate errors using Grafana Traces Drilldown on play.grafana.org. This tutorial guides you through the basic workflow so you can learn by doing.

Give it a try using Grafana Play
Give it a try using Grafana Play

With Grafana Play, you can explore and see how it works, learning from practical examples to accelerate your development. This feature can be seen on the Grafana Play site.

Try it

Because play.grafana.org is a public demo, data changes over time and may reset. If you don’t see data, try refreshing and ensure the time range and data source are correct.

Scenario

You’ve noticed errors in your application and want to investigate which requests are failing and why.

Investigation walkthrough steps

Open Traces Drilldown

  1. Open play.grafana.org.
  2. In the main menu, select Drilldown > Traces.
  3. Traces Drilldown opens with a tracing data source selected.

Choose span data

By default, Traces Drilldown shows Root spans (one span per trace) for accurate, service-level insight. For deeper error investigations, start with Root spans, then switch to All spans to include downstream or internal errors in child spans. Refer to Choose root or full span data for more information.

Select Errors metric

  1. In the Select metric type section, you’ll see three options:
    • Rate: Overall request/span rate
    • Errors: Failed requests/spans
    • Duration: Latency distribution (heatmap)
  2. Click Errors to focus on failed requests.
  3. The view updates to show error-specific tabs: Breakdown, Root cause errors, Comparison, Exceptions, and Errored traces.

View traces with errors

  1. Click the Errored traces tab to see individual traces with errors.
  2. Use the Search field to filter the table by any visible column (for example service name, trace name, or status).
  3. The table shows traces with columns:
    • Start time: When the request occurred
    • Status: Shows error for all traces in this view
    • Trace Service: The service that handled the request (for example, frontend)
    • Trace Name: The operation name (for example, GET /api/quotes, POST /api/pizza)

Examine a specific trace

A trace is a tree of spans that represents a single request as it moves through your services. Each span captures one operation, and child spans nest inside their parent to show the call chain. By understanding the trace structure, you can identify the root cause of errors.

Trace anatomy diagram showing a root span, a child span, and an error span nested in a timeline

  1. Click Open in new tab on a trace row to view the full trace details.
  2. The trace view opens in Explore showing:
    • Trace header: Service name, HTTP method, and status code (for example, 502)
    • Trace ID: The unique identifier for this request
    • Duration: Total time for the request (for example, 250.41ms)
    • Overview timeline: Visual representation of all spans with error spans highlighted
  3. The span timeline shows the sequence of operations:
    • Parent spans (for example, frontend GET /api/quotes)
    • Child spans (for example, HTTP GET)
    • Error spans are highlighted in red/orange.
  4. Use the Filters to focus on:
    • Critical path: The slowest path through the trace
    • Errors: Only spans with errors
    • High latency: Slow spans

Explore error patterns with breakdown

  1. Go back to the Breakdown tab to see error patterns.
  2. The Attributes panel shows available dimensions:
    • resource.service.name: Which services have errors
    • span.name: Which operations are failing
    • span.status: Error status codes
    • span.http.status_code: HTTP response codes
  3. Click on an attribute to break down errors by that dimension.

What you learned

By following this walkthrough, you learned how to:

  • Navigate Traces Drilldown and select the Errors metric.
  • View traces with errors in a table format.
  • Examine individual trace details.
  • Use Breakdown to identify error patterns.

For more advanced investigation techniques like using the Comparison tab and Inspect feature, refer to the Example: Investigate source of errors section.

Next steps

Now that you’ve completed the basic walkthrough:

Use traces with other telemetry signals

Do you have data from other telemetry signals?