Project TechTrends

In this project, I'm the Platform Engineer, tasked with taking an existing Flask application and applying the following practices to make it cloud native. This involves:

1. Increase the Observability of the application. This includes creating API endpoints to gather metrics /metrics as well as the status /healthz. Logging was also done to allow for tracing of errors.
2. Containerizing the application, which involves creating a Dockerfile and a .dockerignore (optional)
3. Creating a Continuous Integration pipeline using Github Actions. Once the main branch is updated, a docker image of the application is built and pushed to Container repository (Docker Hub). Ideally, automated tests should be performed prior to packaging the application, but not required for this exercise.
4. Creating Kubernetes manifests for the application. This allows the application to be deployed in a cluster, in its own namespace.
5. Using a configuration management tool (Helm) to set up different configurations of the application (e.g. sandbox, staging, production).
6. Using a Continuous Delivery tool (ArgoCD) to release different versions of the application.

Changes from the base application

• Used Python 3 for the Docker base image instead of Python 2. This was required as f-strings were used for application logging. F-strings requires Python 3.6 or newer. See https://realpython.com/python-f-strings/
• Used Helm Chart apiVersion: v2, as this is the default value when initialized using helm create. Also, helm lint fails whenever it was set to v1.
• The Vagrantfile was updated as follows:
  • OpenSUSE Leap 15.3
  • Use a different IP for the guest OS from the one suggested as it was unavailable.
  • Installing k3s, helm and argocd on vagrant up using the shell provisioning.
  • Patch the argocd server by changing it from ClusterIP to NodePort, and exposing the relevant port