In our company we have multiple test stages:
- unit and integration tests (CI build)
- application tests (all calls to other systems are mocked)
- system tests (interacting with other systems)
all builds run on TeamCity
As soon as a build fails, one developer has to inform the responsible person which broke the tests. The tests get fixed and run successful again.
The thing is that we assign the role of the "test-fixer" every day to a new developer. He has to take a look at the failed tests and assign them to the person who broke it. This work can be quite annoying since it involves a lot of chatting and stack trace reading.
Is there any tool which makes it easier to manage/assign failed tests? Or whats the best practice to get more stable builds/tests?
We usually run the unit and integration tests before we push to the trunk. But to notice that you broke a application or system test you have to visit the TeamCity page.