[FLINK-25883] The value of DEFAULT_BUNDLE_PROCESSOR_CACHE_SHUTDOWN_THRESHOLD_S is too large

• Type: Bug
• Status: Closed
• Priority: Minor
• Resolution: Fixed
• Fix Version/s: 1.12.8, 1.14.4, 1.15.0, 1.13.7

Environment

Windows, Python 3.8

• Assignee: Dian Fu
• Reporter: Mikhail
• Created: 30/Jan/22 23:55
• Updated: 27/Apr/22 08:28
• Resolved: 14/Feb/22 03:57

Description

In this line, the value of DEFAULT_BUNDLE_PROCESSOR_CACHE_SHUTDOWN_THRESHOLD_S is set to 3153600000. This is more than the default value of threading.TIMEOUT_MAX on Windows Python, which is 4294967. Due to this, "OverflowError: timeout value is too large" error is produced.

Full traceback:

 File "G:\PycharmProjects\PyFlink\venv_from_scratch\lib\site-packages\apache_beam\runners\worker\data_plane.py", line 218, in run
  while not self._finished.wait(next_call - time.time()):
 File "C:\Python38\lib\threading.py", line 558, in wait
  signaled = self._cond.wait(timeout)
 File "C:\Python38\lib\threading.py", line 306, in wait
  gotit = waiter.acquire(True, timeout)
OverflowError: timeout value is too large

Issue Links

• is duplicated by: FLINK-25207 sample pyflink from documentation doesn't work in jupyter notebooks