[ENHANCEMENT] Auto-Instrumentation for OpenAI should be Disabled Off for Auto-Instrumentation for Langchain/Llamaindex

[jlopatec]

We have a lot of instrumentations that use both Auto-Instrumentor for LangChain / Llama and OpenAI. I would like the OpenAI Auto Instrumentor to detect if its running in LangChain / LlamaIndex files. If so disable.

Possibly you could do this by using "inspect" library / looking if we called from LangChain file in frame.filename
# Check if the function is called from a specific file
for frame in inspect.stack():
filename = frame.filename

[mikeldking]

Thanks Jason for the feedback. We have some ideas about making instrumentation composable but langchain itself is currently very difficult. Going to leave this parked here for now.

[dosubot]

Hi, @jlopatec. I'm Dosu, and I'm helping the Arize Phoenix team manage their backlog. I'm marking this issue as stale.

Issue Summary:

• Enhancement proposal to disable OpenAI Auto Instrumentor within LangChain or LlamaIndex files.
• Suggestion to use the "inspect" library to check the call stack for LangChain file calls.
• @mikeldking acknowledged the feedback but noted challenges with LangChain and left the issue unresolved.

Next Steps:

• Please confirm if this issue is still relevant to the latest version of the Arize Phoenix repository. If so, you can keep the discussion open by commenting on the issue.
• If there is no further activity, the issue will be automatically closed in 7 days.

Thank you for your understanding and contribution!