OpenTelemetry Pika Instrumentation

Warnings

• gotcha OpenTelemetry Python SDK and instrumentations must be initialized before the instrumented library (pika) is imported or used. Importing pika before calling `PikaInstrumentor().instrument()` may result in uninstrumented operations.
  Fix: Always place OpenTelemetry SDK setup and instrumentation calls at the very beginning of your application's entry point, before any `import pika` or related code is executed.
• breaking As a beta library (`0.x.y.b0` versioning), `opentelemetry-instrumentation-pika` may introduce breaking changes to its API, configuration, or generated telemetry attributes between minor versions. While efforts are made to align with semantic conventions, these can evolve rapidly.
  Fix: Refer to the `opentelemetry-python-contrib` changelog for specific breaking changes in new releases. Pin exact versions in production environments to avoid unexpected behavior.
• gotcha Automatic instrumentation in multi-process/forking environments (e.g., Gunicorn with multiple workers) can be problematic. OpenTelemetry SDK components, especially `PeriodicExportingMetricReader`, may not function correctly across forked processes due to issues with background threads and locks.
  Fix: For Gunicorn and similar pre-fork servers, consider using a single worker process if automatic metric collection is critical, or use programmatic instrumentation to configure an exporter per worker process. Alternatively, explore zero-code instrumentation via `opentelemetry-instrument` with specific configurations or ensure metrics are pushed via OTLP directly.
• gotcha The instrumentation relies on `pika` being installed and compatible. Recent changes to dependency checks mean that if `pika` is missing or its version is outside the supported range, instrumentation might fail silently or raise an `ImportError` or `DependencyConflictError`.
  Fix: Ensure `pika` is installed and its version meets the requirements specified by `opentelemetry-instrumentation-pika`. Check the `instrumentation_dependencies` method in the source code or `pyproject.toml` for precise version constraints.

Install

• pip install opentelemetry-instrumentation-pika opentelemetry-sdk pika Install core, SDK, and Pika

Imports

• PikaInstrumentor

  from opentelemetry.instrumentation.pika import PikaInstrumentor

Quickstart

This quickstart demonstrates how to instrument the Pika library to automatically generate OpenTelemetry traces. It initializes the OpenTelemetry SDK with a console exporter, instruments Pika, and then performs a basic message publish operation. Ensure a RabbitMQ instance is accessible at `localhost` for this example to run successfully.

import pika
from opentelemetry import trace
from opentelemetry.sdk.resources import Resource
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import ConsoleSpanExporter, SimpleSpanProcessor
from opentelemetry.instrumentation.pika import PikaInstrumentor

# Configure OpenTelemetry SDK
resource = Resource.create({"service.name": "pika-test-app"})
provider = TracerProvider(resource=resource)
processor = SimpleSpanProcessor(ConsoleSpanExporter())
provider.add_span_processor(processor)
trace.set_tracer_provider(provider)

# Instrument Pika
PikaInstrumentor().instrument()

# Pika usage (example: publish a message)
def send_message():
    print("Sending message...")
    connection = pika.BlockingConnection(pika.ConnectionParameters('localhost'))
    channel = connection.channel()
    channel.queue_declare(queue='hello')
    channel.basic_publish(exchange='',
                          routing_key='hello',
                          body=b'Hello World!')
    print("Message sent.")
    connection.close()

if __name__ == "__main__":
    # Ensure RabbitMQ is running or adjust connection parameters
    try:
        send_message()
        print("Traces should be printed to the console.")
    except Exception as e:
        print(f"Error connecting to RabbitMQ or sending message: {e}")
        print("Please ensure RabbitMQ is running and accessible at 'localhost'.")