GCP deployment is down due to billing issue

Live @ http://35.244.39.245

Uses the CD piepline to update on every merge to main branch

Redis uses a persistent volume so even on restart or crash, the data will not be lost

All deployment are configued in a way that during an update, they won't have any down-time

Version deployed at GCP does not uses pvc due to billing issues so to test the same please use minikube

# development - docker
$ docker-compose up dev

# prod - docker
$ docker-compose up prod

# prod - k8s
$ cd k8s && kubectl apply -f .

Docker version will be live @ http://localhost:5000

If using minikube, don't forgot to enable nginx ingress controller using minikube addons enable ingress

/api can be used instead of postman to quick test availble API endpoints.

1. Kraken CI

   • Run jest specs
   • Keeps track of Code Coverage
   • Updates Test Coverage tag

2. Validate Protos
   Checks if there are any breaking changes in the raised PR. Useful to keep track of backward compatibility. ex: PR#3

1. Build and Deploy to GKE
   • Build and upload image to GAR
   • Updates image for the deployment

In addition to the attached K8s resources in this Repo, GCP also uses an static-ip resource in ingress

Available at /metrics

1. http_request_duration_ms : Histogram -> Duration of HTTP requests in ms
2. http_response_status_code: Counter -> Status code of response
3. entries_in_db: Counter -> Number of entries added in db

MIT licensed.