Sign inSign up

cloudreactor/aws-setup-wizard

By cloudreactor

•Updated over 3 years ago

Creates the AWS resources required for running tasks in ECS Fargate and CloudReactor

Image
0

1.8K

cloudreactor/aws-setup-wizard repository overview

⁠cloudreactor-aws-setup-wizard

A command-line wizard to setup customer environments for running tasks managed by CloudReactor

⁠What this can do (pick and choose any or all):

  • Create a VPC, subnets, and a security group for running ECS Fargate tasks
  • Create an ECS cluster
  • Give permissions to CloudReactor to monitor and manage your ECS tasks
  • Create or update Run Environments in CloudReactor so it knows how to run your ECS tasks

⁠Running the wizard

⁠Using Docker

Docker is the recommended way to run the wizard, since it removes the need to install dependencies.

To start, if you haven't already, install Docker Compose on Linux, or Docker Desktop on macOS or Windows.

Once installed, run the Docker daemon.

Next, create a directory somewhere that the wizard can use to save your settings, between runs. For example,

mkdir -p saved_state

Finally run the image:

docker run --rm -it -v $PWD/saved_state:/usr/app/saved_state cloudreactor/aws-setup-wizard

which will use the saved_state subdirectory of the current directory to save settings.

⁠Without Docker (native execution)

First install native python 3.9.x or above. Then clone this repo. In a terminal window, navigate to the repo. Then:

pip install -r requirements.txt
python src/wizard.py

⁠Permissions required / granting access

So that this wizard can create AWS resources for you, it needs the following permissions:

  • Upload CloudFormation stacks
  • Create IAM Roles
  • List ECS clusters, VPCs, subnets, NAT gateways, Elastic IPs, and security groups
  • Create ECS clusters (if using the wizard to create an ECS cluster)
  • Create VPCs, subnets, internet gateways, NAT gateways, route tables, route table associations, VPC endpoints, and security groups (if using the wizard to create a VPC)

You can give the wizard these permissions in a few different ways:

  1. An access key and access secret that you manually enter when prompted by the wizard

  2. Passing the access key and the secret key in the environment variables AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY respectively when you launch the wizard. If you are using Docker to run the wizard:

    docker run --rm -it -v $PWD/saved_state:/usr/app/saved_state -e AWS_ACCESS_KEY_ID=<access_key> -e AWS_SECRET_ACCESS_KEY=<secret_key> cloudreactor/aws-setup-wizard
    
  3. Using your credentials saved in ~/.aws:

    docker run --rm -it -v $PWD/saved_state:/usr/app/saved_state -v $HOME/.aws/:/root/.aws/ -e AWS_PROFILE=<profile_name> cloudreactor/aws-setup-wizard
    

    -e AWS_PROFILE=<profile_name> can be omitted if you just want to use your default AWS profile.

  4. If run from an EC2 instance, the wizard should inherit permissions from the EC2 instance role (not tested yet)

⁠Development

To run possibly modified source code in development

On Linux or macOS, run:

./build.sh

(only needed the first time you get the source code, or whenever you update the source code from the repo)

and then

./wizard.sh

On Windows, run:

.\build.cmd

(only needed the first time you get the source code, or whenever you update the source code from the repo)

and then

.\wizard.cmd

⁠Acknowledgements

Tag summary

Content type

Image

Digest

sha256:3eb5121fd…

Size

65.9 MB

Last updated

over 3 years ago

docker pull cloudreactor/aws-setup-wizard