Hi @Retired_mod - `sys.excepthook` fails to trigger on an unhandled exception. Here's some example code using the logging module. This will write all logging messages to the logfile, except the the ones in `handleException`

import logging
import sys
import datetime
import os
logger = logging.getLogger("log_test")
def handleException(excType, excValue, traceback, logger=logger):
    logger.info("message from exception handler")
    logger.error("Uncaught exception", exc_info=(excType, excValue, traceback))
sys.excepthook = handleException

log_filepath = f'/dbfs/mnt/logs/log_test_{datetime.datetime.now().strftime("%Y%m%d%H%M%S")}.txt'
logging.basicConfig(
    format=f'%(asctime)s - %(name)s - %(levelname)s - %(funcName)s - %(lineno)d - %(message)s',
    level=logging.INFO,
    handlers=[
        logging.FileHandler(log_filepath),
        logging.StreamHandler(stream=sys.stdout)
    ]
)
logger.debug("this is an debug message")
logger.info("this is an info message")
logger.warning("this is an warning message")
logger.critical("this is an critical message")

raise RuntimeError("Test unhandled")

Hey Lozik,

Ran into this myself as well. The reason this doesn't work is because Databricks is using Ipython under the hood.

The following codesnippet creates an exception hook for all exceptions (using the general Exception), it's also possible to specify which exception you wish to add an hook. `exception_handler` was the function I used for handling the exceptions

# Add hook to exception handling
ipy = get_ipython()
ipy.set_custom_exc((Exception,), exception_handler)

Additional documentation can be found here: Document

Good luck!