Our Python agent monitors your Python application to help you identify and solve performance issues. You can also extend your performance monitoring to collect and analyze business data to help you improve the customer experience and make data-driven business decisions. With flexible options for custom instrumentation and APIs, The Python agent offers multiple building blocks to customize the data you need from your app.
Our Python works with a wide variety of web frameworks and hosting mechanisms, including Django, Gunicorn, WSGI, CherryPy, uWSGI, and more. You can also install the Python agent in a Google App Engine flexible environment.
View the big picture of your app:
- Monitor your app's Apdex (user satisfaction).
- Get a high-level summary of your app with Summary page.
- Enable distributed tracing to see activity across an architecture having many services.
- Install Infrastructure monitoring and view detailed server/host data for your app.
Find errors and problems quickly:
- Track key transactions specific to your business.
- Create custom dashboards for important metrics.
- Alert your team when an error or problem occurs before it affects your users.
- View performance after a deployment.
Drill down into performance details:
- Examine code-level transaction traces.
- Examine database query traces.
- Examine error traces.
- Use thread profiler sessions to see detailed stack traces of sampled threads
View logs for your APM and infrastructure data:
Bring your logs and application's data together to make troubleshooting easier and faster. No need to switch to another UI page in New Relic One.
- With logs in context, you can see log messages related to your errors and traces directly in your app's UI.
- You can also see logs in context of your infrastructure data, such as Kubernetes clusters.
Extend agent instrumentation:
Other helpful tools include:
Integrate the Python agent with browser monitoring to gain visibility into end-user browser activity.
Simple scripts and background tasks
Business data analysis with data exploration.
Use the Python agent with our data explorer to organize, query, and visualize your data to answer key questions about application performance and customer experience.
We support a number of web frameworks and libraries right out of the box, including Django, WSGI, and Gunicorn. If you use one of the supported web frameworks, installation is easy. If you use an unsupported framework, the process will involve some additions to your app code and/or web server files.
For a quick and simple install process that will work for the majority of setups, follow these simple steps:
- Download and install the Python package.
- Create config file.
- Integrate the Python agent with your application.
The Python agent also lets you monitor non-web scripts, worker processes, tasks, and functions. The installation process for these non-web transactions is similar to the one used for a web app, with one major difference: instead of going through the standard integration process described in the install instructions, you would manually "wrap" any function you want to monitor. For more information, see Non-web tasks and processes. For instructions on monitoring Celery tasks, see Celery background tasks.
Once you get the agent up and running, some suggested next steps are:
- Explore your data in and get comfortable with the user interface.
- Read our docs on our other observability solutions and the APM page.
- Change your application's name, or other configuration options.
- Learn about setting up custom instrumentation for application activity not monitored by default.
- Consider the Python Telemetry SDK.
After you complete the install process, your data should appear in the APM UI within five minutes. If it does not, use these troubleshooting resources:
- If no data appears, follow these troubleshooting steps.
- If you experience issues when installing or running the Python agent on a new host, test that the package is installed correctly and that it can contact New Relic's data collector service.
- For other problems, see the full list of troubleshooting documentation.