I have a unit test written in mstest which fails when run as part of the solution but passes overall, because the underlying code relies on some static state. This failure is because my tests are being executed in parallel, so other tests clash with this one.
For my testing during development I rely on ncrunch, where I can apply a [Serial]
attribute which prevents that test only from being run in parallel with other tests. Is there something similar I can do for the VS and TFS test runners?
I'm not looking for advice on refactoring my tested code not to use a static method - I'm aware this is not ideal, but am looking for a workaround in this case. I also do not want to disable parallel test execution in general, as there are only one or two which have this requirement.