Capistrano presents a very convinient way to manage your remotely deployed applications and websites. It has initially been developed for Rails deployment. But it also can be very useful for static site deployment.

I’ve been using Capistrano for a couples of months, but I have still some question open. Everything seems to be simple and clear if you grasp main ideas behind the deployment.

One of the buzzing topics is how the source code of your applications gets onto the remote machine. There are two ways for that. The source code can be either uploaded from your local machine, where the Capistrano deployment script is executed, or the remote server (dedicated for the deployment) can fetch the code from an scm repository which should be accessible remotely in this case.

Capistrano supports the idea of deployment strategies. Strictly speaking they are the way, how the source code goes to the remote machine. We can distinguish between local and remote strategies.

Local strategies have only one representant, the copy strategy. Remote strategies split into export, checkout and remote_cache variants.

The above text means: if you’re using a local scm repository which cannot be accessed from the remote machine, you can use only the copy strategy. In that case you use the code:

set :deploy_via, :copy

For more details consider to read the Capistrano source code: https://github.com/capistrano/capistrano/tree/master/lib/capistrano/recipes/deploy/strategy

You may be also interested in this article about deployment of static files: http://f3internet.com/articles/2010/06/18/deploying-static-sites-with-capistrano/