Avoid logging unnecessary stacktraces when shutdown is happening #326
Description
When a map pod is deleted, it logs below critical errors with stack traces:
2026-03-04 02:45:22 CRITICAL UDFError, re-raising the error Traceback (most recent call last): File "/opt/venv/lib/python3.11/site-packages/pynumaflow/mapstreamer/servicer/async_servicer.py", line 52, in MapFn async for msg in global_result_queue.read_iterator(): File "/opt/venv/lib/python3.11/site-packages/pynumaflow/shared/asynciter.py", line 20, in read_iterator item = await self._queue.get() ^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.11/asyncio/queues.py", line 158, in get await getter asyncio.exceptions.CancelledError <-- above error 3 more times --> 2026-03-04 02:45:22 CRITICAL [2026-03-04T02:45:22+0000] CRITICAL [] module=server: [2026-03-04T02:45:22+0000] INFO [] module=server: Killing process: Got exception UDF_EXECUTION_ERROR(udf): CancelledError() 2026-03-04 02:45:22 INFO Killing process: Got exception UDF_EXECUTION_ERROR(udf): CancelledError() /app/entry.sh: line 38: 7 Killed python -m "$EVENT_DRIVEN_APP_MODULE" "$UDF_NAME"This is confusing for the user as this is a normal shutdown and there was no UDF exceptions. We need to catch these exceptions.
Metadata
Metadata
Assignees
Labels
No labels