Realtime chat built with Go, Next, and Typescript

In this project, we will:

1. Develop a Realtime web socket chat using Go, Next, Tailwind CSS, Typescript, and PostgreSQL.
2. Deploy it using Docker Compose.
3. Deploy it using Continuous Delivery to GKE using Github Actions.
4. Monitoring setup using Prometheus. Then, visualize the metrics using Grafana, and finally, set up alerting notifications via email.

Development

• In this section, we look at the development perspective of the project, getting to know its architecture and how it works.
• In simple, it's a three-tier app:
  • The frontend is written in Next.js, tailwind css.
  • The backend with Golang.
  • The database with PostgreSQL.
• To know more about the development process of this app → Development

Deployment - How To Get the Project Working

1- Docker Compose

In This part we:

1. Use docker to containerize our Nextjs frontend, Golang backend and Postgresql database.
2. Make a docker compose file to spin up the project at the click of a button.
3. To see a run in action and To Learn how to get that working → Docker Compose

2- GitHub Actions

In This part we:

1. Configure a domain and DNS server records for our app.
2. Write Kubernetes manifets to our containerized app.
3. Use of Kubenetes gatwayAPI to serve as a http loadbalancer to our app.
4. Write Two Github Actions Pipelines
   • CI for building, testing, linting and artifact saving.
   • CD for deploying our app to GKE.
5. Implement continuous Delivery methodology, beacause maybe we do not want all commits to go for production.
   • We implemented it using github actions Environment review option which gives us two features:
     • a trigger for our production deployment.
     • a review for the changes.
6. To see a run in action and To Learn how to get that working → Github Actions CICD

Monitoring - How To make sure the Project is Working

In This part we:

1. Monitor our application performance using the Golang Prometheus client.
2. Monitor container performance using cAdvisor, and nodes using kube-state-metrics.
3. set up alerting notifications via email.
4. finally, Visualize the metrics using Grafana.