🚀 Feature: Langchain emit error/exception metrics

[smishra-rbx]

Which component is this feature for?

Langchain Instrumentation

🔖 Feature description

Emit error/exception metrics in Langchain instrumentation - similar to OpenAI, Anthropic, Bedrock.

🎤 Why is this feature needed ?

Langchain instrumentation does not seem to be emitting error / exception metrics as part of it's _handle_error method.

Is there a plan to implement this? If yes, would the recommendation be to emit separate metrics for each component type - tool, agent, task, workflow, retriever, etc? It would be nice to have a standardized way for emitting these kinds of metrics.

✌️ How do you aim to achieve this?

Metrics are being emitted for instrumentations like Anthropic, OpenAI and Bedrock. However, Langchain needs a more elaborate implementation here since errors could occur in various components like - tool call, agent, task, workflow, etc. It would be nice to differentiate the emitted metrics based on these components. A simple Counter like Meters.LLM_ANTHROPIC_COMPLETION_EXCEPTIONS might not suffice.

🔄️ Additional Information

No response

👀 Have you spent some time to check if this feature request has been raised before?

• I checked and didn't find similar issue

Are you willing to submit PR?

Yes I am willing to submit a PR!

[nirga]

Thanks @smishra-rbx - we'd love for your contribution :)