Flask Demo

A simple flask demo app deployed into an ubuntu 14.04 docker container, fronted with an nginx container.

Getting Started

These instructions will get you a copy of the project up and running on your local machine using Docker for development and testing purposes. See deployment for notes on how to deploy the project on AWS.

Prerequisites

You will need an AWS account with credentials set up in ~/.aws/credentials as per https://docs.aws.amazon.com/cli/latest/userguide/cli-config-files.html

Assuming you are working on a Mac, apologies to Windows or Linux users currently;

• Download and install docker for mac from here: https://store.docker.com/editions/community/docker-ce-desktop-mac

• Set up Brew https://brew.sh/

• Install the AWS cli brew install awscli

• Install docker-machine for deployment brew install docker-machine

• Set up python env

• virtualenv env ; source ./env/bin/activate

• pip install -r requirements.txt

Local development environment

• Run unit tests FLASK_APP=app/__init__.py python run.py test
• Start app locally FLASK_APP=app/__init__.py python run.py run

Local Docker environment

In order to run the local development environment

• Build and start the Docker containers docker-compose up -d --build

• You can reach the app in a browser at http://localhost

• Check the containers are running using docker-compose ps

russ@russ-mpb:~/Code/flask-demo$ docker-compose ps
       Name                     Command               State           Ports         
------------------------------------------------------------------------------------
flask-demo_app_1     /bin/sh -c python3 run.py  ...   Up      0.0.0.0:5000->5000/tcp
flask-demo_nginx_1   nginx -g daemon off;             Up      0.0.0.0:80->80/tcp    

• Logs can be fetched by running docker-compose logs

• Shut down the stack docker-compose stop

Deployment

In order to deploy this into AWS

• Use docker-machine to provision a Docker host in AWS docker-machine create --driver amazonec2 --amazonec2-region eu-west-2 --amazonec2-instance-type t2.micro --amazonec2-open-port 80 flask-demo

• Once docker-machine reports the machine is ready, configure your docker client to connect to it eval $(docker-machine env flask-demo)

• Build and deploy the container images docker-compose up -d --build

• Test the application is running correctly curl http://$(docker-machine ip flask-demo) or open http://$(docker-machine ip flask-demo) to open it in a browser

Production considerations

The Flask app in this demo is only for development purposes as the built-in server used is not suitable for production. In order to make this production ready we could take steps along the lines of the following.

• Create a docker swarm, or ECS cluster in AWS to enable scalable hosting of containerised application, managed with an infrastructure orchestration tool such as Terraform.
• Use uWSGI in conjunction with nginx rather than Flask's built in server
• Store build docker images in a registry, tagged with appropriate versions
• Set up a CI/CD server such as CircleCI/Jenkins/AWS CodePipeline to handle automated build, test, and deployment of the container images on Git commit
• Configure autoscaling of the application containers and cluster to deal with traffic fluctuations
• Configure log shipping to a log aggregation service such as ELK / Loggly
• Configure monitoring of the application and cluster using tools such as Cloudwatch, Datadog, Newrelic, or AppDynamics
• Offload static assets to S3
• Configure caching - Cloudfront