I am trying to tap into the tracing within crew to allow myself to extract and analyze the thought process that the agents have and the task output. I see that there is a warning when doing so:
Overriding of current TracerProvider is not allowed
Why is this necessary?
As far as i can tell this warning message and tracing the llm thought output are 2 different areas.
This warning message is originating from the opentelemtry package and is getting used for collecting metadata on how crewAI is being used.
For getting more LLM output you can try to set verbose=True when creating Agents and Crew like mentioned in the docs: Agents - CrewAI
For getting all log messages you can set a deeper loglevel like
import logging
logging.basicConfig(level=logging.DEBUG)