# Find out why a run failed

Open the run, then read the failed step: its error, and what it received.

## Open the run

1. Open **Runs**. A failed row says where and why, such as **Failed at Check the order: This order has no total**.
2. Select the row. The run’s page opens on the step that failed.

![A failed run: New order worked, Check the order failed with This order has no total, and Post to the ledger did not run.](https://docs.gorunner.io/assets/img/build/run-page-light.jpg)

The canvas is the version that ran, not your draft. Steps the run didn’t reach are faded. If you move away, **Go to failure** in the top bar brings you back.

Press **Trail** to read the run as a list: the steps that ran, in order, and where it stopped.

## Read the failed step

Under the step’s name, one line says what happened, such as **Failed · after 3 tries · 6.0 s in all · 02:34:50**. The panel has three tabs:

- **Error**: Why it failed. A long error folds; press **Show the whole error**. Below it are the link to the step’s connection if it uses one, **Fix with Keel** for anyone who can edit the workflow, and **Earlier tries** if it was retried. While Keel is asleep, the key stays for someone who can wake it, with **Keel is asleep. Add an AI key to wake it.** under it; anyone else is told whom to ask. Anything the step returned comes last, such as a web request’s response.
- **Input**: What the step received, with every expression already filled in.
- **Settings**: The settings it ran with. You can’t change them here.

![The Error tab of a failed step: the status line, Failed with how long it took and when, then why it failed.](https://docs.gorunner.io/img/build/run-step-failed-light.jpg)

Compare **Input** with what the step expected. An empty field or a wrong path is the usual cause.

**Has this happened before?**

If earlier runs failed at the same step the same way today, the page says how many, with **Open them**. In **Runs**, **Needs you** groups failed runs by cause.

**Why does it say the run started at a step?**

It was a test from that step, such as **Test from** in Runs. The steps above it were filled with earlier test data, not run.

## What the short reason means

The Runs list shortens the error to its kind, such as **the sign-in was refused (HTTP 401)** or **it timed out**. Press **Show the whole error** for the full text. [A run failed](https://docs.gorunner.io/build/a-run-failed) lists each one and its fix.

To move between steps and runs from the keyboard, see [Keyboard shortcuts](https://docs.gorunner.io/build/keyboard-shortcuts#run-page).

## For developers

`GET /api/runs/{id}` returns the run’s `errorMessage` and, in `steps`, each step’s `inputData`, `outputData` and `errorMessage`. See [Get a run](https://docs.gorunner.io/api/runs/get-run).

## Related

- [A run failed](https://docs.gorunner.io/build/a-run-failed): Fixes, by the error on screen
- [Fix failing workflows with Keel](https://docs.gorunner.io/build/fix-failures-with-keel): Let Keel test a fix against the run · 3 min
- [Run again, cancel or delete runs](https://docs.gorunner.io/build/rerun-and-cancel): One run, or many at once · 4 min

Source: https://docs.gorunner.io/build/why-a-run-failed