Increase test timeout for test_handler_traceback_logging

[namannandan]

Description

Regression test test_handler_traceback_logging passes but then fails during teardown in Docker with the following traceback:

================================================================================================================ ERRORS ================================================================================================================
_________________________________________________________________________________________ ERROR at teardown of test_handler_traceback_logging __________________________________________________________________________________________


>       test_utils.torchserve_cleanup()

test/pytest/test_handler_traceback_logging.py:91: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
test/pytest/test_utils.py:110: in torchserve_cleanup
    stop_torchserve()
test/pytest/test_utils.py:93: in stop_torchserve
    subprocess.run(["torchserve", "--stop", "--foreground"])
/usr/lib/python3.9/subprocess.py:507: in run
    stdout, stderr = process.communicate(input, timeout=timeout)
/usr/lib/python3.9/subprocess.py:1126: in communicate
    self.wait()
/usr/lib/python3.9/subprocess.py:1189: in wait
    return self._wait(timeout=timeout)
/usr/lib/python3.9/subprocess.py:1933: in _wait
    (pid, sts) = self._try_wait(0)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

self = <Popen: returncode: -9 args: ['torchserve', '--stop', '--foreground']>, wait_flags = 0

    def _try_wait(self, wait_flags):
        """All callers to this function MUST hold self._waitpid_lock."""
        try:
>           (pid, sts) = os.waitpid(self.pid, wait_flags)
E           Failed: Timeout >60.0s

Turns out that test_utils.torchserve_cleanup() takes longer when running in Docker.

serve/test/pytest/test_utils.py

Lines 109 to 112 in bebab5a

	def torchserve_cleanup():
	stop_torchserve()
	delete_model_store()
	delete_all_snapshots()

Therefore, increasing timeout from 60 seconds to 120 seconds fixes the issue.

Fixes #3106

Type of change

• Bug fix (non-breaking change which fixes an issue)

Feature/Issue validation/testing

• Manual test

$ docker % docker run -it -v /Volumes/workplace/pytorch/serve:/home/serve pytorch/torchserve:ci
========================================================================================================= test session starts ==========================================================================================================
platform linux -- Python 3.9.19, pytest-7.3.1, pluggy-1.5.0
rootdir: /home/serve
configfile: pytest.ini
plugins: timeout-2.3.1, mock-3.14.0, cov-4.1.0
collected 1 item                                                                                                                                                                                                                       

test/pytest/test_handler_traceback_logging.py::test_handler_traceback_logging 
------------------------------------------------------------------------------------------------------------ live log setup ------------------------------------------------------------------------------------------------------------
INFO     root:model_packaging.py:54 Successfully exported model test_model to file /tmp/pytest-of-root/pytest-0/test_model0
DEBUG    urllib3.connectionpool:connectionpool.py:244 Starting new HTTP connection (1): localhost:8081
DEBUG    urllib3.connectionpool:connectionpool.py:549 http://localhost:8081 "POST /models?model_name=test_model&url=test_model.mar&initial_workers=1&synchronous=false&batch_size=1 HTTP/1.1" 202 47
2024-04-24T22:58:02,013 [INFO ] W-9000-test_model_1.0-stdout MODEL_LOG - AssertionErrorPASSED
                                                                                                                                                                                                                           [100%]2024-04-24T22:58:02,014 [INFO ] W-9000-test_model_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - Stopped Scanner - W-9000-test_model_1.0-stdout

---------------------------------------------------------------------------------------------------------- live log teardown -----------------------------------------------------------------------------------------------------------
DEBUG    urllib3.connectionpool:connectionpool.py:244 Starting new HTTP connection (1): localhost:8081
DEBUG    urllib3.connectionpool:connectionpool.py:549 http://localhost:8081 "DELETE /models/test_model HTTP/1.1" 200 52


===================================================================================================== 1 passed in 74.03s (0:01:14) =====================================================================================================

• Docker regression workflow: https://github.com/pytorch/serve/actions/runs/8824491547

[agunapal]

LGTM