Ansible EC2 AMI Stack

Ansible script to create an entire EC2 HTTP stack in AWS using an existing AMI

This role is meant to be run locally. So see the example playbook for how to do that.

Progress

WARNING: This role is a work in progress.

The following currently works:

• create VPC
• add subnet to VPC
• add internet gateway to VPC
• setup route table for VPC and Gateway
• setup security group for HTTP, HTTPS, SSH
• provision EC2 server with an AMI (default Invoke AMI)

The following is not working:

• RDS setup
• S3 setup

Installation

• mkdir -p server-setup/roles
• cd server-setup
• touch playbook.yml
• git clone https://github.com/invokemedia/ansible-ec2-ami-stack roles/invokemedia.ec2-ami-stack

Role Variables

# the application details. used in tags
app_env: stage
app_project: website
app_client: Invoke
# which profile to use in boto
aws_boto_profile: default
# the AMI to use
aws_ami: ami-d4902bb4
# the key-pair to inject into the instance
aws_key_name: user-keyname
# instance type
aws_instance_type: t2.micro
# the region for all the resources
aws_region: us-west-2
# an output of the results of the task
ec2_vars_output: ./ec2-vars.yml

Dependencies

None.

Example Playbook

Here is how you would use the default setup setup.

- hosts: localhost
  connection: local
  gather_facts: True
  vars:
    app_env: stage
    app_project: website
    app_client: Invoke
  roles:
    - { role: invokemedia.ec2-ami-stack }

Then use ansible-playbook playbook.yml -i 'localhost,' --skip-tags "rds,s3" to play.

This role will also create a ec2-vars.yml file in the root of the role. This is used so you can know what the results (ids, names, locations, ips, dns names, etc.) of all the resources are.

License

MIT

Author Information

• Invoke
• [email protected]

References

None.