What are the settings to be set to get Impacted Test results in AzureDev ops for MSTEST

Question

I want to get an Impacted test result in MSTEST but not getting expected result. I have followed all the instructions written here - https://learn.microsoft.com/en-us/azure/devops/pipelines/test/test-impact-analysis?view=azure-devops

This is the log files of VSTS here you can see all the configuration done for Impact Analysis This is test result image where I can not see Impacted results

1. My main branch is "Build Development" and child branch is "Mstest_UT" We have rebased it but still I did not get impacted result as expected.

2. After doing the research I got to know that Impacted test result gets only if all test cases are passed so I did that too but did not get such result.

       [TestMethod]
       public void GetAboutTideContent_Passing_Valid_Data()
       {
           iAboutTideEditorRepository.Setup(x => x.GetAboutTideContent(It.IsAny<ApplicationUser>())).Returns(new AboutTideEditor() { });
           ResponseData<AboutTideEditor> actual = aboutTideService.GetAboutTideContent(It.IsAny<ApplicationUser>());
           Assert.AreEqual(ProcessStatusEnum.Success, actual.Status);
       }
   

I am writing a mock test in MSTEST. I am expecting Impacted test result.

Answer 1

From what I understand from the link you provided for this test you should use this type of test from the start of your project ("growth and maturation off the test" hints towards some kind of deep-learning abilities of the software). If you're kicking in the test halfway, the program might be already locked in commitment of performing particular tests in a certain way (MS stuff remains sometimes having "black box approaches"). If that is the case you should override/reset it and run from the start without having the program or user have selected (detailed) tests. This off-course might set you back for several hours of testing. But consider spending and loosing more time in the search of what goes wrong; it keeps counting an d consuming time if its off the essence to minimize that. Check also the graph provided on the linked page its very informative about the order of actions (e.g. 6).

In your first "black-screen" there is a difference in the parallel setup (consider also below bullets). the black-screen states some dll files are not found in "test assembly". If there is a possibility to run a test-log you might want to check that too to see what typos might have occurred.

From the page:

At present, TIA is not supported for:

• Multi-machine topology (where the test is exercising an app deployed to a different machine)
• Data driven tests
• Test Adapter-specific parallel test execution
• .NET Core
• UWP

In short: reset the whole test and run "fresh" to see if the errors persist.