As testers we must determine impossible assignments. Not just complaining and creating discomfort. Most task can be partially impossible and this also has to be dissected.
You may have:
- A task that is reasonable regarding the project, but not regarding the imposed approach to do the testing.
- Lack of an oracle, where you don’t have the expected behavior defined properly to know what is a bug.
- You don’t have a proper environment or tools to perform the testing in a reliable way
- Misleading or improper documentation
- Lack of a proper cycle in the project. Ex: whatever a developer has done is the correct behavior; the rest is misconfiguration or 3rd party issue
- Difficult environment of some sort in a team
There are a lot of other possible issues that make a project impossible to test or partially impossible. Some situations are resolved with common sense, some need experienced approach.
So before testing is good to determine situations like this and present it.