In this tutorial, we will walk you through the process of instrumenting or tracing a Python FastAPI application using OpenTelemetry and Jaeger. OpenTelemetry is a set of open-source tools that allows you to capture telemetry data in your applications, while Jaeger is a distributed tracing system that helps you monitor and profile the performance of your applications. By instrumenting your FastAPI application with OpenTelemetry and Jaeger, you will be able to effectively trace the lifecycle of requests and pinpoint performance bottlenecks.
Step 1: Install Dependencies
Before we can begin instrumenting our FastAPI application, we need to install the necessary dependencies. You will need to install the following packages:
pip install fastapi
pip install uvicorn
pip install opentelemetry-api
pip install opentelemetry-sdk
pip install opentelemetry-instrumentation
pip install opentelemetry-exporter-jaeger
Step 2: Instrumentation
To instrument your FastAPI application with OpenTelemetry, you will need to initialize the OpenTelemetry SDK and set up the desired exporters. The following code snippet demonstrates how to do this:
from fastapi import FastAPI
from opentelemetry import trace
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import BatchSpanProcessor
from opentelemetry.instrumentation.fastapi import FastAPIMiddleware
from opentelemetry.exporter.jaeger.thrift import JaegerExporter
app = FastAPI()
# Initialize the tracer provider
trace.set_tracer_provider(TracerProvider())
# Create a Jaeger exporter
jaeger_exporter = JaegerExporter(
agent_host_name="localhost",
agent_port=6831,
collector_endpoint="http://localhost:14250/api/traces",
)
# Create a batch span processor
span_processor = BatchSpanProcessor(jaeger_exporter)
# Add the span processor to the tracer provider
trace.get_tracer_provider().add_span_processor(span_processor)
# Instrument the FastAPI app
app.add_middleware(FastAPIMiddleware)In the code snippet above, we first initialize the tracer provider and create a Jaeger exporter. We then set up a batch span processor and add it to the tracer provider. Finally, we instrument the FastAPI app with the FastAPIMiddleware by adding it as a middleware.
Step 3: Starting the application
Once you have instrumented your FastAPI application with OpenTelemetry, you can start the application using the following command:
uvicorn your_app:app –reload
Replace "your_app" with the name of the Python file containing your FastAPI application.
Step 4: Viewing traces in Jaeger
To view traces in Jaeger, you will need to set up a Jaeger instance and configure it to receive traces from your FastAPI application. You can find instructions for setting up Jaeger on the official documentation page: https://www.jaegertracing.io/docs/latest/getting-started/
Once Jaeger is set up, you can navigate to the Jaeger UI and view the traces generated by your FastAPI application. Traces typically include information such as request latency, spans for HTTP handlers, and any exceptions encountered during request processing.
By following the steps outlined in this tutorial, you should now have a better understanding of how to instrument or trace a Python FastAPI application using OpenTelemetry and Jaeger. This setup will help you monitor and debug the performance of your FastAPI applications more effectively.
