Exit course
Chapter 10 of 10 Beginner 7 min read Updated

Knowledge Check

Ten questions on the decisions this course covered: where to run a test, which framework fits, how builds reach devices, and how to read a failure. Answer all of them, then check your answers at once.

1. A team wants feedback on every pull request within ten minutes and a release check on the phones their users own. Which setup fits?

Virtual devices are fast and plentiful, which suits pull requests. Vendor skins, real hardware and performance only show up on real devices, so those runs belong before release, where a slower run is acceptable.

2. Which bug is an iOS Simulator least likely to reveal?

The simulator uses the Mac's memory and CPU, so it does not reproduce a phone's memory limits. Layout, logic and validation bugs show up there just as well as on a device.

3. Android developers want UI tests that run fast, need no explicit waits and live next to the app code. Which framework fits best?

Espresso runs inside the app's process and synchronises with the main thread, so it is fast and needs few waits. Appium needs explicit waits and is slower per command; XCUITest is iOS only; Maestro is black-box and lives outside the app code.

4. An Appium session on a real Android phone fails while installing the app. The same capabilities passed yesterday on a different device. What do you check first?

An install failure happens before any test step runs, so waits and result reporting are irrelevant. A pattern can land on an old OS or a different CPU architecture than the build supports.

5. Where does a CI job get the tb:// id it passes to the Appium tests?

Uploading in the same job ties the test results to the build that was just made and never runs into the 60-day expiry. A hand-maintained id goes stale; the newest upload in the dashboard may belong to someone else's branch.

6. Your plan allows two parallel sessions and a script starts ten Appium sessions at once. What happens?

Sessions beyond the limit wait for a free slot. The risk is on the client side: if the new-session timeout is shorter than the wait, the client reports an error even though the grid would have started the session.

7. An Espresso assertion fails with "Expected: with text: is "1,5" Got: ... text=1.5". The device ran in German. What is the most likely cause?

The view was found and shows 1.5, while the test expects the German decimal comma. Either the app ignores the locale or the test assumes it; both are locale bugs worth fixing deliberately.

8. A test passes when run alone but fails when the whole suite runs on the same device. What is the most likely cause?

Order-dependent results point at shared state. Start each test from a clean app state so it does not depend on what ran before it.

9. Which TestingBot CLI exit code should make a pipeline retry the job automatically, rather than report failing tests?

1 is a CLI or infrastructure error, where a retry can help. 2 means the tests ran and failed, which a retry would only hide; 0 is success.

10. A new app supports Android 10 and up and iOS 17 and up, and you have budget for six real devices in the release matrix. How do you choose them?

The matrix exists to represent your users. The oldest supported versions catch compatibility bugs; newest flagships alone or random free devices leave the real user base untested.

Where to go from here#

You now have a working picture of mobile app testing: which device answers which question, which framework fits which team, and how a run on a cloud device is set up, scaled and debugged. To go deeper in one framework:

To earn the certificate, pass the exam and run the practical task on TestingBot.

You finished Mobile App Testing: From First Test to CI

Put what you learned into practice. Run your tests on 6100+ real browsers and physical iOS and Android devices.