Both tests are concerned with handling exceptions from the child process.
This might not be an actual bug, due to this note, but I’m not sure about it. If it isn’t a bug, we could maybe check whether we are able to reliably suppress the stack trace printing during the tests.
This is happening on the master branch, but I can also reproduce it locally. The logs snippet below is from the master branch, see here for the full build.
+ coverage run --source app --omit '*_test.py' --module unittest discover --pattern '*_test.py' app.log
.........................Exception in thread Thread-1:
.Traceback (most recent call last):
File "/home/circleci/.pyenv/versions/3.9.17/lib/python3.9/threading.py", line 980, in _bootstrap_inner
. self.run()
File "/home/circleci/.pyenv/versions/3.9.17/lib/python3.9/threading.py", line 917, in run
self._target(*self._args, **self._kwargs)
File "/home/circleci/project/app/execute_test.py", line 31, in raise_exception
raise Exception('Child exception')
Exception: Child exception
.Process ProcessWithResult-2:
Traceback (most recent call last):
File "/home/circleci/.pyenv/versions/3.9.17/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
self.run()
File "/home/circleci/project/app/execute.py", line 37, in run
result.return_value = self._target(*self._args, **self._kwargs)
File "/home/circleci/project/app/execute_test.py", line 31, in raise_exception
raise Exception('Child exception')
Exception: Child exception
...........................................................................................................................................................
----------------------------------------------------------------------
Ran 183 tests in 0.802s
In
execute_test.py
, there are two test cases which are producing error output while running, without making the tests fail, though:test_process_with_result_child_exception
~test_execute_with_timeout_child_exception
test_background_thread_ignores_function_exception
Both tests are concerned with handling exceptions from the child process.
This might not be an actual bug, due to this note, but I’m not sure about it. If it isn’t a bug, we could maybe check whether we are able to reliably suppress the stack trace printing during the tests.
This is happening on the master branch, but I can also reproduce it locally. The logs snippet below is from the master branch, see here for the full build.