Wubot

Balanced Hubot installation

Installing hubot-scripts to a destination

make install DEST='${ hubot_dir }/scripts will copy all the scripts from the hubot-scripts directory to your target destination.

Deploying

This project has been engineered as a proof-of-concept for scalable deployments with AWS's Elastic Beanstalk for Docker. Hence, the easiest way to deploy this will be using AWS's eb cli version 3.0 or higher.

• Creating an environment (from scratch)

  • eb create wubot -ip wubot
  • See .ebextensions section below

• Deploying an update

  • add a script to the hubot-scripts folder
  • commit and push git repository
  • eb deploy

• How to add a dependency to the docker image?

  • TODO (need to rebuild the docker image..)

.ebextensions

TODO

01_balanced_specific_aws_vpc.config

TODO

02_setup.config

TODO

03_app_environment.config

This file represents the equivalent of a docker -e flag.

Scripting

The hubot-scripts directory contains all the hubot scripts that will ultimately end up in the ${ hubot_dir }/scripts directory. Just add your coffee scripts into the hubot-scripts.

Read this: https://github.com/github/hubot/blob/master/docs/scripting.md#scripting for more information

Prototyping with the "shell adapter"

• https://railsmachine.com/articles/2012/05/23/building-a-bot-with-hubot/

Testing your script

TODO

Hacking

The Vagrantfile attached will use a docker-host machine as a host to build a docker container that runs the wubot installation.

vagrant ssh will ssh into the docker container

You may override anything by passing down the variable via command-lines, for example:

• HUBOT_BOTNAME will override the bot's name
• HUBOT_HIPCHAT_ROOMS will override what rooms the bot joins
• HUBOT_ADAPTER will override the default hipchat adapter

Re-using a different base image

Currently, the Vagrantfile uses a build_dir setting:

config.vm.define 'wubot' do |wubot|
  # Sync the project directory using rsync
  wubot.vm.provider 'docker' do |d|
     d.build_dir = './docker/'
     # ...
  end
end

You can read about this here: http://docs.vagrantup.com/v2/docker/configuration.html

If you want to iterate on an existing image, you must change the Dockerfile to use:

FROM balanced/wubot:1.0.2

LICENSE

MIT/BSD dual-licensed