​
Overview

This guide will walk you through setting up Langtrace’s Python SDK with the OpenTelemetry (OTEL) Collector to send traces to Langtrace’s Cloud or a self-hosted setup.

​
Configurations

​
OTEL Collector

Refer to the OTEL configuration page for details on running and configuring the OpenTelemetry Collector with a custom configuration file.

OTEL Configuration

Ensure that you configure the Langtrace API token in the OTEL collector configuration file before proceeding.

​
Python SDK

To send traces to the OTEL Collector, install the required OpenTelemetry exporter package:

pip install opentelemetry-exporter-otlp-proto-http

​
Implementation

Initialize the Langtrace SDK with a custom remote exporter using the OTLPSpanExporter to send traces.

langtrace_otel.py
from langtrace_python_sdk import langtrace
from openai import OpenAI
from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter


# Configure the OTLP exporter to use the correct endpoint
otlp_endpoint = "http://localhost:4318/v1/traces"
otlp_exporter = OTLPSpanExporter(
    endpoint=otlp_endpoint,
    headers={"Content-Type": "application/json"}
)
langtrace.init(custom_remote_exporter=otlp_exporter, batch=False)


def chat_with_openai():
    client = OpenAI()
    messages = [
      {
          "role": "system",
          "content": "How many states of matter are there?",
      },
    ]
    chat_completion = client.chat.completions.create(
        messages=messages,
        stream=False,
        model="gpt-3.5-turbo",
    )
    print(chat_completion.choices[0].message.content)


chat_with_openai()

Run the script:

python langtrace_otel.py

​
Conclusion

By following this guide, you will have integrated OpenTelemetry support into your Python project, allowing you to send traces to an OpenTelemetry backend, including Langtrace’s Cloud or a self-hosted setup.

TypeScript SDKPOST Project