Infrastructure for Multi-AZ web application

Tools used:

Docker - webapp application Packer - create AMI AWS CLI - AWS command line interface Terraform - deploy IaaC

Setup steps

1. Clone this repository to your local environment.

2. AWS CLI setup:

1. Install AWS CLI. 
    1. Command: bash brew install aws cli 
2. Configure AWS CLI
    1. Command: aws configure 
    2. Insert your IAM user AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, region “eu-west-1” and output type.

3. Packer setup:

1. Install Packer
    1. Command: brew install packer
2. Configure Packer to build AMI. It is required to have AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY and IAM user with appropriate permissions.
    1. Command: export AWS_ACCESS_KEY_ID={your AWS_ACCESS_KEY_ID}
    2. Command: export AWS_SECRET_ACCESS_KEY={your AWS_SECRET_ACCESS_KEY}
3. Subscribe for CentOS AMI if require : https://aws.amazon.com/marketplace/pp?sku=aw0evgkw8e5c1q413zgy5pjce
4. Validate and build web app AMI.
    1. Command: packer validate webapp.js
    2. Command: packer build webapp.js

4. Provision AWS infrastructure using Terraform:

1. Deploy S3 bucket to store Terraform states.
    1. Switch to terraform/state/ directory.
    2. Change “s3_name“ variable value in variables.tf to something unique. That will be re-used in further steps.
    3. Change value of attribute “bucket” in terraform.tf to the same what was used in 4.1.2. for “s3_name”.
    4. Create S3 bucket:
        1. Command: terraform init
        2. Command: terraform apply
    5. Remove “_postinit” from terraform.tf_postinit file.
    6. Move terraform states from local to s3
        1. Command: terraform init
2. Deploy multi-az infrastructure
    1. Switch to terraform/webapp directory.
    2. Change value of attribute “bucket” in terraform.tf to the same what was used in 4.1.2 for “s3_name”.
    3. Initialize Terraform project.
        1. Command: terraform init
    4. Create “terraform-key-webapp” Key Pair via AWS Console in Ireland region.
    5. Conduct a dry run to validate terraform code
        1. Command: terraform plan
    6. Provision AWS resources via terraform:
        1. Command: terraform apply
        2. Wait for AWS resources creation.
    7. Wait for a couple of minutes for EC2 instances initialization and later copy “webapp_elb_dns_name” value to your browser to check if web app is available.

AutoScaling mechanisms

CloudWatch Scaling CPU metric based:

Every two minutes check for ASG average CPU utilization.
Scale-out in case of CPU Utilization is higher or equal 40% => add 1 instance to ASG
Scale-in in case of CPU Utilization is lower or equal 10% => remove 1 instance from ASG

List of pre-defined scheduling rules for autoscaling:

Peak time     => Starting from 1pm, min and desired capacity 6, max 9
Off peak time => Starting from 9pm, min and desired capacity 3, max 6
Night time    => Starting from 1am, min and desired capacity 1, max 6
Morning time  => Starting from 7am, min and desired capacity 3, max 6