ieeesb-mcmaster-website

Website for McMaster

Test Project Locally

Run:

docker compose up

Simulate Production Environment

To simulate the production environment locally (note this requires kubernetes installed as well as a kubernetes service provider) You need to do the following steps

1. Create a Github PAT token (project settings, under development settings, tokens, create classic token, read:package permission only enabled)
2. Run the following kubernetes commands in the root of the project (ensure that both loadbalancing and storage is enabled on microk8s if using that)

kubectl create secret docker-registry ghcr-secret \
  --docker-server=ghcr.io \
  --docker-username=${Github user name} \
  --docker-password=${Github personal access token} \
  --docker-email=${Github email}
kubectl apply -f=postgres.dev.volumes.yaml -f=postgres.dev.yaml -f=webapp.yaml

If the image has changed due to someone merging branches with main, Run the following:

kubectl delete -f=webapp.yaml
kubectl apply -f=webapp.yaml

If you want to clean postgres as well then

kubectl delete -f=postgres.dev.yaml
kubectl apply -f=postgres.dev.yaml

Production

Setup Kubernetes Environment on production

To setup the environment on production, run the following command in the machine. For now, we are not using persistent volumes and pvcs for production since I had trouble setting it up and syncing it, but that is something I'll look into when the PI's are online again.

kubectl apply -f=postgres.yaml -f=webapp.yaml

When a feature has been added to main (via merging)

Since all builds are automated on the main branch with Github Actions, it will generate a docker image in Github Packages. Everytime this occurs, you would need to SSH into the PIs and perform the following commands:

kubectl delete -f=webapp.yaml
kubectl apply -f=webapp.yaml