Inploy was born as an option to setup and deploy Rails applications in an easier and smarter way.
It automatically executes common tasks like migrate the database, install gems, package the assets, clean the cache, notify Hoptoad, restart the server, etc. This means you will not have a big deploy script calld ‘recipe’ like with other tools, only a simple deploy script with a few lines.
It uses Git to version the deploys and to pull or push only the modifications in each one. This means you will not have a lot of folders in the server (only one) with duplicated code and that ech deploy will be very fast.
It is very flexible, meaning you can create templates specifing how specific tasks must behave. As today, there are templates for Locaweb, for Rails 3 using git pull and Rails 3 using git push.
It works by default with Passenger, but you can specify another servers like Mongrel, Thin and Unicorn.
You can execute Inploy from a remote machine or from the server machine, meaning anynone can go to the deployment path and with only one command update the code and execute all the tasks.
As Inploy is very flexible, now you can also use it to execute an installation shell script in a server from an URL.
As a plugin:
script/plugin install git://github.com/dcrec1/inploy.git
As a gem:
sudo gem install inploy
Please remember that when used as a gem, Inploy should be available in the deploy servers.
Rails 2.3.x note
When using rails 2.3.x, you should include the follow line into you Rakefile:
##### Rakefile load "tasks/inploy.rake"
When using Inploy with Rake, there are five tasks:
- inploy:remote:install from=url
- connects to a list of servers and executes the script from url
- connects to a list of servers - clones a git repository - executes rake inploy:local:setup
- inploy:remote:update or inploy:up
- connects to a list of servers - executes inploy:local:update
- connects to a list of servers - reads the content of url specified by from=url without downloading the file and executes it
- creates directories db and tmp/pids - copies config/.[example|sample|template] files to config/ - creates the database - executes init.sh file case it exists - inits and updates the git submodules - installs gems - migrates the database - updates the crontab from Whenever if being used - cleans the cache, default tmp/cache - cleans cached assets in public/assets if Jammit is installed - starts the Delayed Job worker if script/delayed_job exist - parses less files if more:parse task exists - parses compass files if config/initializers/compass.rb exists - parses coffeescript files with Barista if barista:brew task exist - package the assets if asset:packager:build_all task exists - notifies Hoptoad about the deploy, specifing the revision, repository and environment - notifies NewRelic RPM about the deploy - restarts the server
- pulls the repository - inits and updates the git submodules - copies config/.[example|sample|template] files to config/ - installs gems - migrates the database - updates the crontab from Whenever if being used - cleans the cache, default tmp/cache - cleans cached assets in public/assets if Jammit is installed - restarts the Delayed Job worker if script/delayed_job exists - parses less files if more:parse task exists - parses compass files if config/initializers/compass.rb exists - parses coffeescript files with Barista if barista:brew task exist - package the assets if asset:packager:build_all task exists - notifies Hoptoad about the deploy, specifing the revision, repository and environment - notifies NewRelic RPM about the deploy - restarts the server
If you don’t want to execute Rake, there are a few shortcuts:
inploy => inploy update => rake inploy:remote:update inploy setup => rake inploy:remote:setup inploy install from=url => rake inploy:remote:install from=url
When using inploy as a command, it looks first for config/deploy.rb and then for deploy.rb .
Create a config/deploy.rb file and configure it something like this:
application = "signal" repository = 'git://github.com/dcrec1/signal.git' hosts = ['hooters', 'geni'] # OPTIONALS before_restarting_server do rake "important:task" run "important_command" end path = '/opt' # default /var/local/apps user = 'dcrec1' # default deploy ssh_opts = '-A' # default empty branch = 'new_version' # default master environment = 'staging' # default production sudo = true # default false cache_dirs = ['public/cache', 'tmp/cache'] # default ['public/cache'] skip_steps = ['install_gems', 'clear_cache'] # default  app_folder = 'project_folder' # default empty login_shell = true # default false bundler_opts = '--binstubs' # default '--deployment --without development test cucumber'
bundler_path option has been removed. To migrate your existing config/deploy.rb file simply replace
bundler_path = '/path/to/gems' with
bundler_opts = '--path /path/to/gems'
You can skip some steps when running a Rake task, just add “skip_steps” into your rake command line or add “skip_steps” to your deploy.rb to always skip those steps.
Currently the steps available to skip are:
- copy_sample_files: skip copying config/.[example|sample|template] files to config/ - install_gems: skip rake gems:install - migrate_database: skip rake db:migrate - clear_cache: skip removing cache directories - update_crontab: skip updating the crontab with Whenever
- Usage (params are comma separated):
rake inploy:remote:update skip_steps=install_gems,migrate_database
Currently, the following three hooks are available:
- before_git (not called before initial git clone) - after_git - after_setup - before_restarting_server - after_restarting_server
To show a “down for maintenance” page during the deployment, use the before_git and after_restarting_server hooks.
If you need another hook, please fork the project and implement it. Or submit an issue, it’s very easy.
say method to show messages issued by your hooks.
# Show a simple message say 'restart custom daemons' # Show a message (blank output lines are suppressed) say 'revert local changes' do |output| output << `git checkout . 2>&1` end
Inploy is brought to you by:
- Diego Carrion
- Douglas Campos
- Kristopher Murata
- Vitor Pellegrino
- Josh Owens
- Josef Pospíšil
- Andy Shen
- Erik Dahlstrand
- Carlos Brando
- Thomas Ritz
- Sven Schwyn
(The MIT License)
Copyright © 2010
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the ‘Software’), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED ‘AS IS’, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.