Search

Troubleshooting 'No Available ML Data' Errors

Overview

The "No available ML data" error indicates that the machine learning model lacks sufficient information to execute a test step. This often occurs when a test case has been recently copied or edited without being re-executed, or when a preceding step fails due to an application issue.

How It Works

When troubleshooting this error, it's important to investigate systematically, as the error may be a symptom of a different, underlying problem. Follow these steps to diagnose and resolve the issue:

  1. Check for Preceding Application Errors: Review the test execution results carefully. An application-related error, such as an unexpected validation message appearing on the page, can occur in a step before the "No available ML data" error is reported. This earlier failure prevents the test from proceeding correctly, leading to the ML data issue on a subsequent step. Resolve any application errors first.
  2. Run Copied Test Cases Before Editing: When a test case is copied, the machine learning data from the original test is not transferred. You must execute the newly copied test case at least once to generate a baseline of ML data. Editing a copied test before its initial successful execution is a common cause of "No available ML data" errors.
  3. Perform a Local Edit: If the application is stable and the error persists, perform a Local Edit on the failing step. Re-selecting the element or re-performing the action can generate the necessary ML data for the model to learn from. After the local edit, save the test and execute it again.
  4. Ensure Element Visibility: Sometimes, an element is not available because it is outside the visible area of the page. If you suspect this is the case, consider adding a Scroll action before the failing step to bring the target element into view.

Limitations

The effectiveness of these troubleshooting steps depends on a stable application environment. If the application under test has inconsistent behavior or bugs, it can cause a variety of test failures that mask the true root cause. It is difficult to resolve ML data issues if the test fails for different application-related reasons across multiple executions.

Related Info

If the issue persists after following these steps, please create a support ticket for each distinct issue. This allows them to be investigated independently and more effectively. Provide links to the test case and the specific execution results showing the problem.