Real Estate

This application is collecting SeLoger real-estate products on demand and store them to a database. The user requires an account to login and start importing products. He/She can then list or search for those products later on.

Architecture

The micro-services architeture is described in the below diagram:

flowchart LR
   USER --:80--> FE([front-end])
   USER --:8080--> AUTH([auth svc])
   USER --:8081--> PG([pg-admin])
   subgraph services
      FE -.-> GTW([api-gateway]) & AUTH
      GTW -.:8080.-> LOG([logger svc]) -.:27017.-> DBM[(mongo \n DB)]
      GTW -.:8080.-> PROD([product svc]) -.:5432.-> DB[(postgres \n DB)]
      GTW -.:8080.-> AUTH -.:5432.-> DB
      PG -.:5432.-> DB
   end

Loading

Front End

Vue.js Single Page Application (SPA) using Vuetify 3 component library, Vite tooling and Pinia store management.

API Gateway

KrakenD merge and map routes of all micro-services APIs and secure endpoints using the authentication service.

Authentication Service

Keycloak handles OAuth2 authentication using OpenId worflow. It delivers a JWT token to the client once it is authenticated.

Product Service

The product micro-service is a REST API handling CRUD operations on real-estate products. The database can be initialized using the import endpoint which connects to the SeLoger API and eventually store the results in the database.

Logger Service

The logger micro-service is a REST API handling a single endpoint /log. The MongoDB database logs all actions happening within the solution.

Deployment

Local using Compose

Install docker, make and compose and run the following command:

make up_build

Browse localhost:8080 to configure the authentication portal:

• Open Administration Console
• Login using admin/admin username and password
• Make sure the Invokify realm is selected
• Open Users menu and click add user button
• Create a new user by filling in Email, First & Last Name and check the box Email Verified, then click save.
• Open Credentials tab and set a new password (uncheck the box Temporary)

Browse localhost:80 to start the application.

Cloud using Kubernetes

The .gitlab-ci.yml configures gitlab pipeline for you, in order to build the docker images and then create resources in your Google GKE cluster.

Do not forget to change the context name (nicolashvallee/go-real-estate:gke-agent-prod) with your own.

Before that, follow the below instructions:

1. Create the K8S cluster

   Follow instructions to create a Google GKE cluster or simply activate the docker one.

2. Deploy the containers

   Deployment files are defined in ./deployment/kubernetes folder, and user can appply resources using the following commands:

   kubectl apply -f ./deployment/kubernetes/registry-credentials.yml
   kubectl apply -f ./deployment/kubernetes/postgres.yml
   kubectl apply -f ./deployment/kubernetes/pgadmin.yml
   kubectl apply -f ./deployment/kubernetes/keycloak.yml
   kubectl apply -f ./deployment/kubernetes/backend.yml