Flask-Redis-Postgres-K8S-App

a simple flask app with a postgres backend. We have used kubernetes and helm apis to deploy the application on a local kubernetes installation (kubernetes on docker desktop)

Run using kubernetes

to run the app using kubernetes APIs (commands):

1. create secret for postgres admin password kubectl create secret generic postgres-secrets --from-literal=password=postgres-admin-password

2. create secret for Azure Container Registry login kubectl create secret docker-registry acrregistrykey --docker-server=link-to-registry --docker-username=username --docker-password=access-key

3. update:

   • container image location at containers.image key in flask.deployment.yml file
   • the path key for postgres-init volume in the postgresql.deployment.yml file to the init script (it should be to src/database/db_create.sql file)
   • the path key for postgrespv in the postgres.pv.yml file and point to a folder to save the postgres database dumps/backups. This location can be used to restore the postgres db in case the container crashes.

4. Apply the k8s config files kubectl apply -f k8s/

Run using helm

1. create a secretvalues.yml file at the root of the charts (k8shlmcharts folder)
2. populate the secretvalues in the file as :

   postgresdb:
     password: postgres-password
     init_script_pth: "/Complete/path/to/file/src/database/db_create.sql"
     postgres_db_backup: "/Complete/path/to/folder/postgresdbvolume"
   
   flask-app:
     acrloginkey: "credentials to login to the docker registry where the images is loaded (dockerhub, acr etc.)"
     registryname: registry+image-name:tag
   

3. run command from the root (k8shelmcharts folder) of the helm charts directory helm install flaskapp . -f secretvalues.yml

To Do:

1. setup on actual kubernetes cluster
2. setup ingress services
3. setup a cloud storage as a volume
4. setup deployment pipelines to push to kubernetes cluster