Adding test for a valid config update after an invalid config

[fabian-moessner]

No description provided.

[ekneg54]

This does not reproduces the failure. Please have a look deeper.
unittests for the refresh interval can be found in tests/unit/test_runner.py. Consider starting there to go deeper into testing and bug reproduction.

[ekneg54]

as described in

Logprep/logprep/util/configuration.py

Lines 434 to 435 in 0068804

	version: str = field(
	validator=validators.instance_of(str), converter=str, default="unset", eq=True

the version attribute is of type "str". So setting it to the string "invalid" doesn't result in an invalid configuration.

[ekneg54]

as defined here

Logprep/logprep/runner.py

Lines 120 to 129 in 0068804

	@_config_refresh_interval.setter
	def _config_refresh_interval(self, value: int | None) -> None:
	"""Set the configuration refresh interval in seconds."""
	if value is None:
	self._configuration.config_refresh_interval = None
	elif value <= 5:
	self._configuration.config_refresh_interval = 5
	else:
	self._configuration.config_refresh_interval = value

the minimum for a config refresh interval should be 5 Seconds. If you are able to set it to 4 Seconds we might have a new bug.

[ekneg54]

sleep actions in test leads to flaking tests because you never know if a test run on a slower system does not needs some seconds more. It is better to wait for some condition in the logs for example.