Cross-browser validation is an important part of quality. In mabl, all tests run on Chrome by default. Depending on your workspace's pricing plan, you can also configure tests to run on Firefox, Internet Explorer 11, Safari and Edge to ensure your website is working as expected across different browsers.

📘

Note

Mabl does not support Edge in IE mode at this time.

This guide outlines how to configure browser settings for your tests.

Configuring browser settings

At the plan level

On the Plans page, open the Plan Details page for the plan you would like to configure. Then take the following steps:

1. Click on the Edit icon in the top left corner to open the Edit plan page.
2. Scroll down to "Device settings."
3. Choose which browsers you would like to run the plan on.

4. Click the "Save plan" button at the bottom of the page.

🚧

Adding browsers will increase runs

When a plan or test is configured to run on multiple browsers, the number of test runs increases proportionately. For example, if you configure a plan with eight tests to run on three different browsers, each plan run will trigger up to 24 tests (assuming the plan run isn't stopped early by a failed test).

At the test level

You can also configure browser settings for ad hoc cloud runs. Open the Test Details page for the test you would like to run and take the following steps:

1. Click "Run test." The "Ad hoc run" panel will appear.
2. Select "Cloud run."
3. Choose the browsers you would like to run the test on.
4. Add any other configurations as needed.
5. Click "Start run" to trigger the test run(s).

Limitations

Safari duration times

When running your tests on Safari, you may encounter some scheduling latency as a result of mabl's use of real Apple hardware for all Safari tests. Thus, there are also some limitations on parallel execution to provide the best testing experience to all customers. In a situation affected by latency, tests will be executed in the order they are requested and as quickly as possible.

It is also important to note that the test execution duration time includes the time it takes from clicking run/trigger to when the OSX VM is spun up, which may look like a discrepancy between the time the steps took to execute vs. the full duration.