Skip to content

Project 4 on the 'Monolith to Microservices' course at Udacity CloudDeveloper ND

License

Notifications You must be signed in to change notification settings

maloufde/udagram-microservices-course

Repository files navigation

UDAGRAM MICROSERVICES COURSE Build Status

Introduction

This Repository contains the code for Project 4 of the Udacity Cloud-Developer Nanodegree on Monolith to Microservices decomposition. It implements a simple Instagram-like App to enable the user to register, login and upload images.

The Frontend is currently publicly accessible at:

The Rest-Api (e.g. feed) is deployed to:

Overview of a Kubernetes deployment

This is an Overview of the app as it is deployed to a Kubernetes cluster on AWS EC2 instances.

Udagram Kubernetes Deployment

The app is assembled of 3 domain specific parts:

  • an SPA-frontend based on IONIC,
  • a backend feed microservice that will store and retrieve the images on an AWS S3 bucket,
  • a backend user microservice for registering and logging into the app.

An additional reverse-proxy will route api-requests to the appropriate microservice, based on the URL.

Last but not least there is a deployment part with information and config files to launch the app on Docker and Kubernetes.

In order to review or try out this project easily, it includes all these components as subfolders, knowing that this would not be the preferred structure for team development on microservices. Please have a look to the appropriate subdirectory for further details on each part.

Basic Requirements

As outlined in the course you'll need to setup different components:

  • AWS S3 bucket, aka 'AWS media bucket'.
  • AWS credentials for an IAM user with sufficient rights to access the media bucket
  • PostgreSQL database, accessible from your environment
  • Docker installation
  • Docker-Compose installation (>= v1.23.2)

Prepare environment

With basic requirements fulfilled, you should have ~/.aws/credentialsin place and be able to configure the following environment variables:

  • UDCD_AWS_CREDENTIALS The path to the AWS credentials, e.g. /users/bill/.aws or c:/users/bill/.aws (1)
  • UDCD_AWS_PROFILE: The AWS profile to take credentials from, e.g. default
  • UDCD_AWS_REGION: The AWS region, e.g. us-east-1
  • UDCD_AWS_MEDIA_BUCKET: The AWS S3 bucket where the images are stored
  • UDCD_CORS_ALLOW: The origin to be allowed to access the backend services from Javascript
  • UDCD_JWT_SECRET: The secret for JWT authentication
  • UDCD_POSTGRES_HOST: The public domain name of the PostgreSQL Database instance
  • UDCD_POSTGRES_DATABASE: The name of the PostgreSQL Database
  • UDCD_POSTGRES_PASSWORD: The password of the PostgreSQL Database
  • UDCD_POSTGRES_USERNAME: The name of the PostgreSQL User

Notes:

  1. On Docker Desktop for Windows, the bind mounted directory must be enabled as Shared Drive in Docker Settings.

Build and Run all images with docker-compose

Change the API-URL in udacity-c3-frontend\src\environment.ts from
apiHost: 'http://api.udagram.christian-malouf.de/api/v0/' to apiHost: 'http://localhost:8080/api/v0/'

# build
cd udacity-c3-deployment/docker
docker-compose build

# run    
docker-compose up -d    

Check status with docker-compose ps, tail logs with docker-compose log -f.

Open the frontend in your Browser: http://localhost:8100

Deployment on Kubernetes Cluster

Kubernetes resources are configured in subdirectory udacity-c3-deployment/kubernetes. See scripts setup-kubectl.sh and update-kubernetes-namespace.sh for details. These scripts are used on Travis-CI to populate the app into one namespace per branch. You can select the Travis-CI passed-Image to see the previous build.

Collecting Logfiles in CloudWatch

Container Logfiles are forwarded to AWS CloudWatch, see README-CloudWatch for further details.

About

Project 4 on the 'Monolith to Microservices' course at Udacity CloudDeveloper ND

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published