Flask_app_deployed_to_AWS_using_AMI

---

Summary

This is a RESTful APIs based web application that allowed users to register themselves and upload thier files in any format to s3 bucket built with Flask, PostgreSQL, and deployed on AWS

• EC2 instances are built on a custom AMI using packer
• Setting up the network and creation of resources is automated with Cloud formation, aws cli and shell scripts
• Instances are autoscaled with ELB to handle the web traffic
• Created a serverless application to send a mail to user once he registers. SES and SNS
• The application is automaticaly tested using pytest and deployed using Github workflow once the pull request is merged.

Architecture Diagram

Tools and Technologies

   

Infrastructure	VPC, ELB, EC2, Route53, Cloud formation, Shell, Packer
Webapp	Python, Flask, PostgreSQL, DynamoDB
CI/CD	Github Worflow
Alerting and logging	statsd, Cloud Watch, SNS, SES, Lambda
Security	KMS and EBS for encryption of RDS and EC2, and SSL certificate for domain name using Route53 and DNS.

Infrastructure-setup

• Create the networking setup using AWS CloudFormation and AWS CLI
• Create the required IAM policies and users
• Setup Load Balancers, Route53, DynamoDB, SNS, SES, RDS

Webapp

• The Web application is developed using Flask framework that uses the REST architecture
• Secured the application with Basic authentication to retrieve user information
• Storing the files uplaoded by users in S3 bucket and retrieving as an object.
• Generating [verification URL] with expiration of 3 minutes using DynamoDB

Build Instructions

Pre-Requisites: Need to have postman installed

• Clone this repository into the local system
• Go to the folder Flask_app_deployed_to_AWS_using_AMI /webapp
• Download all the packages listed in requirements.txt using pip
• Run WebappApplication in your local server by running the command flask run app.py

CI/CD

• After pushing the code to github the workflow will run the unit test case to check the working of app.
• After merging the code the github workflow will create a custom AMI and push it into EC2 instance.
• The artifact generated is stored in S3 bucket and deployed to an autoscaling group.

Auto scaling groups

• Created auto scaling groups to scale the EC2 instances for the application to handle the webtraffic and keep the costs low when traffic is low
• Created cloud watch alarms to scale up and scale down the EC2 instances

Serverless computing

• Created a pub/sub system with SNS and lambda function
• When the user registers, a message is published to the SNS topic.
• The lambda function checks for the entry of the email in DynamoDB if it has no entry then it inserts a record with a TTL of 5 minutes and sends the notification to the user with SES ![alt text]

Packer

• Implemented CI to build out an AMI and share it between organization on AWS
• Created provisioners and bootstrapped the RDS instance using shell script