AWS Autoscaling

A basic example to provision a resilient application using auto scaling group.

Config

Directory structure

modules/
  app/
  vpc/
stage/
  terraform.tfvars
  ...
<prod>/
  <clone stage directory>

You can create a new directory for your other environments to keep the terraform.tfvars, user_data.sh, tfstate, etc., separate.

Setup

git clone git@github.com:ergurjeet/aws-autoscaling.git
cd aws-autoscaling/stage
terraform init
terraform plan
terraform apply

Exceptions

• "Unable to find remote state": Create an empty state file on S3

{
  "version": 4,
  "outputs": {},
  "resources": []
}

and then run terraform init

Simulate cpu load

SSH into a node

sudo su
amazon-linux-extras install epel -y
yum install stress -y
stress --cpu  2 --timeout 300