/ECI-aygo-lab1

Create a webapp to save texts using docker and AWS.

Primary LanguageJavaScript

ECI-aygo-lab1

This is a small laboratory of the course "AYGO" at Escuela Colombiana de Ingeniería Julio Garavito and part of the Informatics's master degree.

Objectives

Create an e2e web application that saves strings and returns the 10 most recently created. Also I implemted an application load balancer to balance the load when saving the texts.

The project was build using docker-compose, go and react, deployed on an EC2 instance.

Demo: http://ec2-34-209-136-168.us-west-2.compute.amazonaws.com/

How to run locally

$ git clone https://github.com/Lufedi/ECI-aygo-lab1
$ docker-compose build
$ docker-compose up -d

Warning If you are running this on your local machine ignore this, otherwise if you are running this on a remote server, change the backend url to your server public domain or IP here. (EC2 domain i.e.)

Project structure

  • app-lb-round-robin: Application load balancer using the RR algorithm. go
  • aygolabone: Backend service that saves and retrieves the strings.go
  • aygolaboneclient: Webapp to send and read the strings. reactjs
  • docker-compose: Docker tool to orchestrate multiple containers in one project

Arquitecture and design

The architecture is as described in the image bellow.

Overview

Three backend services (aygolabone) are running on containers at port 4000, those services domains where registered in the app-lb-round-robin project manually*, which is a reversed proxy and a gateway at the same time, the app-lb-round-robin picks one of the three services using the round robin algorithm and forwards the request to the selected service.

This was achieved using an already implemented Proxy tool called SingleHostReverseProxy

(*) this step can be avoided using the built-in service discovery tool from docker and docker-compose, but one goal of this laboratory is study the RR load balancing algorithm.

The API and the webapp

The Webapp is a simple react app that calls the exposed API to ger and save the texts. The API signatures are:

GET /api/text

POST api/test { value: "value" }

For more details on the API go to request.http

The database

Mongo was used to save the texts on a single collection with two fields, value and creation_date

Deployment

The application was deployed in a EC2 Instance (manually)

Demo

This is a picture of the 3 backends services receiving the requests with an even distribution

Containers you should see after running docker-compose

TODO

  • Base project running in containers
  • Implement health check to discard offline services
  • Document code
  • Validate input in the webapp form
  • Use Nginx