External dependencies make a test flaky, while the other integration test is not. By separating them, we can keep the existing integration test in our daily canary runs while excluding the end-to-end tests, thereby avoiding false positives
Edit: updated to fully exclude the end-to-end tests from the CI. So to test them, they now need to be run manually
External dependencies make a test flaky, while the other integration test is not. By separating them, we can keep the existing integration test in our daily canary runs while excluding the end-to-end tests, thereby avoiding false positives
Edit: updated to fully exclude the end-to-end tests from the CI. So to test them, they now need to be run manually