Wait, retry, assert vs require conditions in component tests not necessarily meaningful, investigate more robust conditions.

[entlein]

Description

CTs sometimes fail for (mostly its timing issues), and the tests are red, even though the functionality is green. This is wasteful

Environment

github hosted runners, see config in .github (stable for last year)

Steps To Reproduce

run the CTs, see that some sometimes fail

Expected behavior

CTs should fail when what the tests assert is broken

Actual Behavior

CTs sometimes fail due to flakyness and rerunning the tests makes them green.

Additional context

It might be simply helpful to categorize which types of flakyness we currently have. A zoology of test-devils, so to say.

This bug is not high prio, just me being impatient with waiting for test runs

BTW: the benchmark has similar issues