Flask-Datadog

0.1.4 · maintenance · verified Thu Apr 16

Flask-Datadog is a micro-framework extension for Flask applications, providing access to DogStatsD for custom metrics. Last updated in 2017, it offers a simple wrapper around the `datadog` library's StatsD and API clients. For comprehensive application performance monitoring (APM), tracing, and logging with Flask, Datadog's officially supported `ddtrace` library is the current recommended solution.

Common errors

```
ModuleNotFoundError: No module named 'flask.ext.datadog'
```
cause You are attempting to import `flask_datadog` using the deprecated `flask.ext` namespace, which was removed in Flask 1.0.

fix Change your import statement from `from flask.ext.datadog import ...` to `from flask_datadog import ...`.
```
ConnectionRefusedError: [Errno 111] Connection refused
```
cause The Datadog Agent's DogStatsD server is not running or is not accessible at the configured `STATSD_HOST` and `STATSD_PORT`.

fix Ensure the Datadog Agent is installed and running, and that its DogStatsD port (default 8125) is open and reachable from your Flask application. Verify `app.config['STATSD_HOST']` and `app.config['STATSD_PORT']` are correctly set.

Warnings

deprecated The `flask.ext` import mechanism, commonly shown in older `flask-datadog` examples, is deprecated since Flask 0.11 and removed in Flask 1.0. Attempting to use it with modern Flask will result in an `ImportError` or `ModuleNotFoundError`.
Fix: Update imports from `from flask.ext.datadog import ...` to `from flask_datadog import ...`.
gotcha `flask-datadog` is an older library (last updated 2017) primarily for StatsD metrics. For comprehensive Datadog integration, including APM, distributed tracing, and advanced logging, Datadog officially recommends using the `ddtrace` library with its automated instrumentation for Flask applications.
Fix: Consider migrating to `ddtrace` for new projects or enhanced monitoring. If sticking with `flask-datadog`, understand its limited scope to DogStatsD and basic API calls, and that it may not be compatible with newer Flask features or Python versions without manual intervention.

Install

pip install Flask-Datadog Install latest version

Imports

API
```
from flask_datadog import API
```
StatsD
wrong:
```
from flask.ext.datadog import StatsD
```
correct:
```
from flask_datadog import StatsD
```
The 'flask.ext' import path was deprecated in Flask 0.11 and removed in Flask 1.0. Direct import from 'flask_datadog' is required for modern Flask versions.

Quickstart

This quickstart initializes Flask-Datadog for both StatsD metric submission and Datadog API interactions. It demonstrates incrementing a custom metric on a successful route and another on an error route. Ensure `DATADOG_STATSD_HOST`, `DATADOG_STATSD_PORT`, `DATADOG_API_KEY`, and `DATADOG_APP_KEY` environment variables are set for full functionality, though StatsD will default to localhost:8125.

from flask import Flask
from flask_datadog import API, StatsD
import os

app = Flask(__name__)

# Configure Datadog API and StatsD settings
app.config['STATSD_HOST'] = os.environ.get('DATADOG_STATSD_HOST', 'localhost')
app.config['STATSD_PORT'] = int(os.environ.get('DATADOG_STATSD_PORT', '8125'))
app.config['DATADOG_API_KEY'] = os.environ.get('DATADOG_API_KEY', '')
app.config['DATADOG_APP_KEY'] = os.environ.get('DATADOG_APP_KEY', '')

statsd = StatsD(app)
dogapi = API(app)

@app.route('/')
def hello():
    statsd.increment('my_flask_app.requests')
    return 'Hello, Datadog!'

@app.route('/error')
def error_route():
    statsd.increment('my_flask_app.errors')
    # Example of API usage (requires valid API/App keys)
    # if dogapi.api_key and dogapi.app_key:
    #     dogapi.Event.create('An error occurred in Flask app', 
    #                        tags=['env:dev', 'service:my-app'])
    raise Exception('This is a test error!')

if __name__ == '__main__':
    # Ensure the Datadog Agent is running and accessible at STATSD_HOST:STATSD_PORT
    # For API calls, ensure DATADOG_API_KEY and DATADOG_APP_KEY are set.
    app.run(debug=True)

view raw JSON →