The test runner dialog automatically shows the next available test case in the current test set. The test cases are executed in the order they appear in the test set.

For test sets that are not set to sequential, which enforces the order of test cases you can navigate to the next, previous, or last test case in the test set using the arrows on the navigation bar in the Test test Runner runner dialog.

Enter the test case number in the input box on the navigation and press enter.

Click the search icon next to the navigation bar to bring up the Choose a Test... dialog. The dialog lists all of the not-completed test cases. These test cases can be filtered and any of them can be selected as your next test case.

The test runner dialog shows the test cases which are not executed yet. If you have already run Test "X" within the test run, that test case is not available in the test runner.

You can stop and close the test runner dialog any time during the run and come back later. The execution is continued at the next not-yet-run test case, which is where you left off, unless somebody else has run the same test.

After each step is completed, the test runner dialog automatically saves the results in the background. This avoids data loss in case the browser crashes or other issues are encountered.

Multiple testers can work on the same test set run simultaneously. The system automatically takes care of concurrency. If test case "X" is being run by another user then that test case is not available to the next tester. No work is duplicated. If the same user account runs the same test set from two different browsers (multiple sessions), they do not get the same test case. This prevents accidental overwriting of other user’s work.

If a test case has parameters, the test case is run as many times as the number of parameters. The test runner does not go to the next test case until all parameters are completed. Some parameters can be skipped. For more details, see Test Parameterization.

Typically, only test cases which are in the Accepted state are run by the test runner, but this can be configured. For more information, see Running Only Accepted or All Test Cases?.