Separation of Concerns - Regressiontests and reusable Tests in Xray

[Daniela Maubach]

Hey Community - it's me again 

I have the following question:
We have test cases that are written in general terms and can therefore also be executed regressively.

Now we checked the “Separation of concerns” setting so that the request status on the boards remains meaningful.
My expectation was that with this setting the other results would be displayed but not changed, but they changed to the status of the actual test execution.
Of course now i can specify a custom result for each request, but is it correct that no historical events are taken?

Example:
T1 was used for R1(passed), R2 (fail) and R3 (passed).
R4 failed.  

My Expectation: These status are shown...

Now in the test execution all 4 requirements are set to failed.

Of course, I can manually reset these to the original results, but to do this I would have to open each individual task and research the test results.

Am I making a mistake in my thinking, does Xray act like this or have I overlooked a setting? Or is it an error?
 

• What are your best practices for reusable testing? 
  Creating new test cases every time is too high due to the regulatory effort with a 4-eye review.
• ⁠Out of interest: Can someone tell me whether separation of concerts is now available in the cloud?

best regards 

Daniela

[Manuel]

Hi @Daniela Maubach ,

1. The setting for separation of concerns nay be misleading. By default, in Xray if you have a test linked to several requirements, if the test fails all those requirements will be marked as not being ready (NOK). It you uncheck it, then you can specify how the result is reflected on each linked requirement.. so unchecking the setting is like enabling the separation of concerns. It depenfs on the perspective but the behavior right now is that. For your use case I would recommend to make test cases that are focusing on validating the scope of each requirement so you can clearly identify what's their goal. If you have a test that is testing too much then whenever it fails you cannot understand exactly what is actually impacting. Also, smaller tests are easier to understand and execute

2. Separation of concerns setting is not available on cloud, at least right now

[Daniela Maubach]

Screenshot of the test execution details i mean