Skip to content

Latest commit

 

History

History
109 lines (73 loc) · 4.04 KB

INSTALL.md

File metadata and controls

109 lines (73 loc) · 4.04 KB

Setup For Localhost Development

1. Generate new config files

After cloning the project from github. Run:

bundle install --without production --path vendor/bundle

The following command makes it easy to get started with the configurations required to run the application. Run this command in the directory containing the project to generate configs with reasonable defaults:

rails g config

This generates the following files:

Dockerfile
docker-compose.yml
.esaas-engagements-config.json
config/application.yml

The above files are .gitignored to avoid checking in secret credentials.

To view the list of available options for the config generator run the following command: rails g config --help

1b. (Optional) Edit the config cache and re-run config generator

You can now edit the .esaas-engagements-config.json with custom values eg. setup a SendGrid API key after setting one up from their website. After customizing the .esaas-engagements-config.json you can re-run rails g config to regenerate the config files with these new values. .esaas-engagements-config.json files stores your config values and uses them to regenerate config/application.yml and your docker files.

2. Start application

You can now start the application locally either on "bare-metal" or inside docker.

2a. "Bare metal" Approach

For this approach your laptop need to have dependencies such as nodejs and phantomjs (for testing) installed either using Homebrew(MacOS), Chocolatey(PC) or your Linux package manager.

bundle install --path vendor/bundle && rails s

2b. Docker Approach

For this approach you do not need to have dependencies on your host machine installed manually.

Build the docker image:

docker-compose build

Run migrations:

docker-compose run rails-server bundle exec rake db:migrate

Seed the database:

docker-compose run rails-server bundle exec rake db:seed

Since there is volume mapping, you only need to run migrations and the seed once, not unless there is a new migration or changes to seed.rb.

Spawn the docker network:

docker-compose up -d

The docker machine that is spawned has volume mapping so changes in code should be reflect immediately. However, changes to the Gemfile* files will require re-building the docker image with docker-compose build.

Stop the docker network:

docker-compose down

To run tests in your docker machine:

docker-compose run -e "RAILS_ENV=test" rails-server bundle exec rake test

To ssh into the docker instance, run the following command after docker-compose up -d:

docker exec -it rails-server /bin/bash

3.Deploying to heroku.

Read the documentation for working with Rails 4 on Heroku.

If you want to deploy to heroku you should ideally re-clone the repository in a new folder different from your local development folder. Re-run rails g config --no-docker in the new clone. Edit the .esaas-engagements.json with values you want to use in production eg. if you have a SENDGRID key you want to use for production, overwrite the json key for the key with appropriate values.

Now run figaro heroku:set -e production -a YOUR+HEROKU+APP+NAME+HERE.

Replace YOUR+HEROKU+APP+NAME+HERE with example-app if your heroku app is at example-app.herokuapp.com.

4. Set up simple Continuous Deployment(CD)

Step 1: Set up travis.yml

Read the documentation for Heroku Deployment with Travis.

deploy:
  provider: heroku
  api_key:
    secure: "[You Key Here]"
  app: [Your app name on Heroku]
  on:
    repo: "[Your Repo name and path]"
  • For API Key: generate short-term API Key with travis encrypt $(heroku auth:create)
  • For App Name: use Heroku deployment app name
  • For Repo Name: use your repo name with github id (ex: my-github/esaas-engagements)

Step 2: Configure Heroku

  • Go to your Heroku App page -> Deploy
    • Deployment method -> Connect to Github with your repo
    • Automatic deploys -> Enable "Auto deployment" and "Wait for CI Pass"