Hi!
A few months ago we switched from WinRM to PyWinRM to connect rundeck to our windows nodes, where we run multiple python scripts.
This has resulted on some jobs showing up as failed on rundeck, even though they concluded successfully on the windows machine, given the fact that PyWinRM can't process characters outside of UTF-8, unlike WinRM. So, when a python script tries to log something outside of UTF-8, the job fails.
This an output example:
06/02/2025 12:46:09 - INFO - #####################START#####################
06/02/2025 12:46:09 - INFO - Execution mode REPROCESS
06/02/2025 12:46:09 - INFO - Reprocessing mode
[ERROR ] 'utf-8' codec can't decode byte 0xe3 in position 17162: invalid continuation byte (winrm-exec.py:362)[root]
[ERROR ] 'utf-8' codec can't decode byte 0xe3 in position 446: invalid continuation byte (winrm-exec.py:362)[root]
[ERROR ] 'utf-8' codec can't decode byte 0xe3 in position 429: invalid continuation byte (winrm-exec.py:362)[root]
And this only started happening after the switch to PyWinRM. As a temp fix, we are changing the logs to not include "à" or "õ" or those kind of characters (which are quite frequent in portuguese), but we were wondering... Is this fixable? Is this a setting on rundeck or on the node definition?
Thanks in advance for any help.