Django ToDo List

In this task you will enhance a todo list web application by adding a /metrics endpoint that returns the number of GET and POST requests in a Prometheus-compatible format.

To do this task, you will need:

• CSS | Skeleton
• JS | jQuery

Explore

Try it out by installing the requirements (the following commands work only with Python 3.8 and higher, due to Django 4):

pip install -r requirements.txt

Create a database schema:

python manage.py migrate

And then start the server (default is http://localhost:8000):

python manage.py runserver

Now you can browse the API or start on the landing page.

Task

Follow the steps below to complete the task:

1. Fork the repository to your GitHub account.

2. Create a new /metrics endpoint (not api/metrics):

   • Modify the application code to include a /metrics endpoint.
   • Ensure it returns the number of GET and POST requests in a Prometheus-compatible format.

3. Add Prometheus library:

   • Include prometheus_client in the application's requirements.
   • Ensure it's installed during the Docker build process.

4. Cluster setup:

   • Use kind to spin up a cluster from a cluster.yml configuration file.

5. Pull the kube-prometheus-stack:

   • Prometheus

   helm pull prometheus-community/kube-prometheus-stack --version <version_number> --untar

6. Install the kube-prometheus-stack Helm chart:

   helm install kube-prometheus-stack ./kube-prometheus-stack

   • Note: If you encounter issues deploying on Ubuntu, particularly related to the admission controller, find the admissionWebhooks section in the values.yaml and set:

   admissionWebhooks:
     enabled: false

7. Modify the values.yaml file in the todoapp Helm chart:

      serviceMonitor:
        enabled: true
        labels: {}
        interval: 10s
        path: /metrics
        port: http

8. Configure your todoapp Helm chart:

   • In the Helm chart's templates directory, create a new file named servicemonitor.yaml with the necessary configuration.
   • Ensure your app's service.yaml exposes the metrics port and has the necessary labels for the ServiceMonitor to select.
   • Perform port forwarding to your application service to expose the metrics endpoint:

     kubectl port-forward svc/<your-service-name> 8080:<app-port>

9. Apply Helm Chart:

   • Apply the Helm chart to the cluster.
   • Verify that the ServiceMonitor has been created.

10. Verify Prometheus scraping:

• Check Prometheus dashboard for the target status to ensure it's scraping metrics from the todoapp.

11. Create Grafana dashboard:

• Open the Grafana UI and click on the + icon in the sidebar to create a new dashboard.
• Click Add new panel to start configuring your first metric visualization.

12. Visualize total HTTP requests:

    • Set up a panel titled Total HTTP Requests.
    • Use the query to visualize the total number of HTTP requests:

      sum(rate(<your_custom_metric>[5m])) by (method)
      

    • This query will show the rate of HTTP requests per second, averaged over the past 5 minutes, broken down by method (GET or POST).
    • Choose a visualization type such as Graph, Bar Gauge, or Stat to represent this data.

13. Visualize HTTP requests creation time:

    • Add a panel titled HTTP Requests Creation Time.
    • Use the query to visualize the creation time of the requests:

      <your_custom_metric>
      

    • This metric represents the time when the HTTP request counters were created or reset, which can be useful for identifying when the application was restarted.
    • Choose a Singlestat or Stat visualization for this metric.

14. Customize panel settings:

    • Adjust panel settings such as axes, legend, and thresholds as desired.
    • Utilize Grafana’s functions for formatting the display.

15. Submit a PR with your changes and attach screenshots of your Grafana dashboard for validation on the specified platform.