Closing because I’m not sure this is an issue, if you are convinced that this is really a bug, please feel free to re-open the issue and add more information (your versions (Ruby, Cap, etc), your Capfile, your logs, and relevant sections out of your Gemfile)

Otherwise support is done via the mailing list (https://groups.google.com/forum/#!forum/capistrano) or at StackOverflow (http://stackoverflow.com/questions/tagged/capistrano) with questions tagged capistrano.

FTR, had the same issue, found the solution on StackOverflow.

https://coderwall.com/p/4k1lja/fixing-capistrano-3-deployments-after-a-repository-change

I've run into this several times. The solution in each case is just to nuke the repo directory. This isn't an issue when it is just one server, but in one case I'm deploying to 6 servers. However, changing the repo url is rather uncommon, so it is most just an annoyance.

For a workaround, try this: https://rubygems.org/gems/capistrano-git-delete_repo

As I was reading the comments in the link above, I noted that someone suggested this:

namespace :git do
  task :update_repo_url do
    on roles(:all) do
      within repo_path do
        execute :git, 'remote', 'set-url', 'origin', fetch(:repo_url)
      end
    end
  end
end

This might be worth making a standard part of the Git deployment workflow. I doubt that it would take long on deploy, and might solve this issue.

I can't see what version of Git that was added to - I'll check the changelogs - https://git-scm.com/docs/git-remote

@stamm Thank you! I deleted dir repo and it solved my problem!