Capistrano uses a strictly defined directory hierarchy on each remote server to organise the source code and other deployment-related data. The root path of this structure can be defined with the configuration variable :deploy_to.

Assuming your config/deploy.rb contains this:

set :deploy_to, '/var/www/my_app_name'

Then inspecting the directories inside /var/www/my_app_name looks like this:

├── current -> /var/www/my_app_name/releases/20150120114500/
├── releases
│   ├── 20150080072500
│   ├── 20150090083000
│   ├── 20150100093500
│   ├── 20150110104000
│   └── 20150120114500
├── repo
│   └── <VCS related data>
├── revisions.log
└── shared
    └── <linked_files and linked_dirs>
  • current is a symlink pointing to the latest release. This symlink is updated at the end of a successful deployment. If the deployment fails in any step the current symlink still points to the old release.

  • releases holds all deployments in a timestamped folder. These folders are the target of the current symlink.

  • repo holds the version control system configured. In case of a git repository the content will be a raw git repository (e.g. objects, refs, etc.).

  • revisions.log is used to log every deploy or rollback. Each entry is timestamped and the executing user (:local_user, defaulting to the local username) is listed. Depending on your VCS data like branch names or revision numbers are listed as well.

  • shared contains the linked_files and linked_dirs which are symlinked into each release. This data persists across deployments and releases. It should be used for things like database configuration files and static and persistent user storage handed over from one release to the next.

The application is completely contained within the path of :deploy_to. If you plan on deploying multiple applications to the same server, simply choose a different :deploy_to path.

Fork me on GitHub