Basic Two-Tier AWS Architecture

This provides a template for running a simple two-tier architecture on Amazon Web services. The premise is that you have stateless app servers running behind an ELB serving traffic.

To simplify the example, this intentionally ignores deploying and getting your application onto the servers. However, you could do so either via provisioners and a configuration management tool, or by pre-baking configured AMIs with Packer.

This example will also create a new EC2 Key Pair in the specified AWS Region. The key name and path to the public key must be specified via the terraform command vars.

After you run terraform apply on this configuration, it will automatically output the DNS address of the ELB. After your instance registers, this should respond with the default nginx web page.

To run, configure your AWS provider as described in https://www.terraform.io/docs/providers/aws/index.html.

Run with a command like this:

terraform apply -var 'key_name={your_aws_key_name}' \
   -var 'public_key_path={location_of_your_key_in_your_local_machine}'

For example:

terraform apply -var 'key_name=terraform' -var 'public_key_path=/Users/jsmith/.ssh/terraform.pub'

Alternatively to using -var with each command, the terraform.template.tfvars file can be copied to terraform.tfvars and updated.