Modifica

Condividi tramite

Tutorial: Use the .NET Aspire dashboard with Python apps

  • Articolo

The .NET Aspire dashboard provides a great user experience for viewing telemetry, and is available as a standalone container image that can be used with any OpenTelemetry-enabled app. In this article, you'll learn how to:

  • Start the .NET Aspire dashboard in standalone mode.
  • Use the .NET Aspire dashboard with a Python app.

Prerequisites

To complete this tutorial, you need the following:

  • Docker or Podman.
    • You can use an alternative container runtime, but the commands in this article are for Docker.
  • Python 3.9 or higher locally installed.
  • A sample application.

Sample application

This tutorial can be completed using either Flask, Django, or FastAPI. A sample application in each framework is provided to help you follow along with this tutorial. Download or clone the sample application to your local workstation.

git clone https://github.com/Azure-Samples/msdocs-python-flask-webapp-quickstart

To run the application locally:

  1. Go to the application folder:

    cd msdocs-python-flask-webapp-quickstart

  2. Create a virtual environment for the app:

    py -m venv .venv
.\.venv\Scripts\Activate.ps1

  3. Install the dependencies:

    pip install -r requirements.txt

  4. Run the app:

    flask run

  5. Browse to the sample application at http://localhost:5000 in a web browser.

    Screenshot of the Flask app running locally in a browser

Adding OpenTelemetry

To use the .NET Aspire dashboard with your Python app, you need to install the OpenTelemetry SDK and exporter. The OpenTelemetry SDK provides the API for instrumenting your application, and the exporter sends telemetry data to the .NET Aspire dashboard.

  1. Install the OpenTelemetry SDK and exporter:

    pip install opentelemetry-api opentelemetry-sdk opentelemetry-exporter-otlp-proto-grpc

  2. Add a new file to your application called otlp_tracing.py and add the following code:

    import logging
from opentelemetry import metrics, trace

from opentelemetry._logs import set_logger_provider
from opentelemetry.exporter.otlp.proto.grpc._log_exporter import (
    OTLPLogExporter,
)
from opentelemetry.exporter.otlp.proto.grpc.metric_exporter import OTLPMetricExporter
from opentelemetry.exporter.otlp.proto.grpc.trace_exporter import OTLPSpanExporter
from opentelemetry.sdk._logs import LoggerProvider, LoggingHandler
from opentelemetry.sdk._logs.export import BatchLogRecordProcessor
from opentelemetry.sdk.metrics import MeterProvider
from opentelemetry.sdk.metrics.export import PeriodicExportingMetricReader
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import BatchSpanProcessor

def configure_oltp_grpc_tracing(
    endpoint: str = None
) -> trace.Tracer:
    # Configure Tracing
    traceProvider = TracerProvider()
    processor = BatchSpanProcessor(OTLPSpanExporter(endpoint=endpoint))
    traceProvider.add_span_processor(processor)
    trace.set_tracer_provider(traceProvider)

    # Configure Metrics
    reader = PeriodicExportingMetricReader(OTLPMetricExporter(endpoint=endpoint))
    meterProvider = MeterProvider(metric_readers=[reader])
    metrics.set_meter_provider(meterProvider)

    # Configure Logging
    logger_provider = LoggerProvider()
    set_logger_provider(logger_provider)

    exporter = OTLPLogExporter(endpoint=endpoint)
    logger_provider.add_log_record_processor(BatchLogRecordProcessor(exporter))
    handler = LoggingHandler(level=logging.NOTSET, logger_provider=logger_provider)
    handler.setFormatter(logging.Formatter("Python: %(message)s"))

    # Attach OTLP handler to root logger
    logging.getLogger().addHandler(handler)

    tracer = trace.get_tracer(__name__)
    return tracer

  3. Update your application (app.py for Flask, main.py for FastAPI) to include the imports and call the configure_oltp_grpc_tracing function:

    import logging
from otlp_tracing import configure_otel_otlp

logging.basicConfig(level=logging.INFO)
tracer = configure_otel_otlp()
logger = logging.getLogger(__name__)

  4. Replace the print calls with logger.info calls in your application.

  5. Restart your application.

Framework Specific Instrumentation

This instrumentation has only focused on adding OpenTelemetry to our code. For more detailed instrumentation, you can use the OpenTelemetry Instrumentation packages for the specific frameworks that you are using.

  1. Install the Flask instrumentation package:

    pip install opentelemetry-instrumentation-flask

  2. Add the following code to your application:

    from opentelemetry.instrumentation.flask import FlaskInstrumentor

# add this line after configure_otel_otlp() call
FlaskInstrumentor().instrument()

Start the Aspire dashboard

To start the Aspire dashboard in standalone mode, run the following Docker command:

docker run --rm -it -p 18888:18888 -p 4317:18889 --name aspire-dashboard \
    mcr.microsoft.com/dotnet/aspire-dashboard:9.0

In the Docker logs, the endpoint and key for the dashboard are displayed. Copy the key and navigate to http://localhost:18888 in a web browser. Enter the key to log in to the dashboard.

View Structured Logs

Navigate around the Python application, and you'll see structured logs in the Aspire dashboard. The structured logs page displays logs from your application, and you can filter and search the logs.

Screenshot of the Aspire dashboard showing the Python application logs

Next steps

You have successfully used the .NET Aspire dashboard with a Python application. To learn more about the .NET Aspire dashboard, see the Aspire dashboard overview and how to orchestrate a Python application with the .NET Aspire app host.