CloudSeeder is an extension to the wonderful troposphere library that allows for the easy creation of robust AWS CloudFormation custom resources. These custom resources can be used to extend the capabilities of AWS CloudFormation to include new services, configuration sources, or any bespoke functionality.

At the current time, this library is in an exploratory state and probably doesn't work completely if at all. This notice will be removed once the framework is stabilized and plugin support is added.

TODO

These steps use the AWS CLI to upload and deploy the stack artifacts, and assume a POSIX shell environment. $DEPLOYMENT_BUCKET should be set beforehand to a bucket you own in the account. You can add flags like --profile and --region to the AWS CLI commands to customize the account and region the stack will be deployed to.

python3 -m virtualenv venv
. venv/bin/activate
pip install cloudseeder
python3 -m cloudseeder.deploy
python3 -m awscli cloudformation package \
    --template-file cloudseeder-template.json \
    --s3-bucket $DEPLOYMENT_BUCKET \
    --output-template-file cloudseeder-template-packaged.yaml
python3 -m awscli cloudformation deploy \
    --capabilities CAPABILITY_IAM \
    --template-file cloudseeder-template-packaged.yaml \
    --stack-name CloudSeeder
deactivate

Note: CloudSeeder must be installed into a virtualenv (or pyvenv) in order to create its deployment artifacts. Attempting to run cloudseeder.deploy outside of a virtualenv will not function.

TODO

This will use setuptools entry points or some similar mechanism to allow the code bundle to include resource classes not defined within this package.

Unit tests can be run via python setup.py test. They currently pass on both Python 2.7 and Python 3.5.

Integration tests are a TODO.