similar to: Deployment problem with Capistrano

For those who want use switchtower to deploy to a remote server that does not have access to the main subversion repository, here is a task I just hacked up to use switchtower to deploy to my servers by copying a local directory to the servers, rather than have the servers use svn checkout. I actually create my local directory using an svn export, but you can do whatever you like here, and just

I''m finding the :before_task and :after_task facilities in Capistrano very useful. For example, I needed to add a shared directory to my system. All I had to do was add a few lines to deploy.rb: desc "Add shared usr_img directory to shared" task :after_setup do run <<-CMD mkdir -p -m 775 #{shared_path}/user_img CMD end desc "Link shared usr_img directory

My app is set up without database.yml in version control. I created a shared_path/config dir, and put database.yml in it. After I update the code, I want to link the app''s database.yml file to the shared config. So I added this task desc ''Copy the database config'' task :after_update_code, :roles => :app do run "ln -s #{shared_path}/config/database.yml

Hi all, I want to love capistrano and I''m so close. My error is occuring while running the update_code task... (apologies for the shell spew) ************************************************************************* * executing "rm -rf /home/henster29/sites/glitter/releases/20060316143447/log /home/henster29/sites/glitter/releases/20060316143447/public/system &&\n ln

How does one configure capistrano for moving local file to your remote server? In the docs it says that one cannot use file:// so what''s the configuration step for resolving this issue? Thanks in advance, -Conrad -------------- next part -------------- An HTML attachment was scrubbed... URL: http://wrath.rubyonrails.org/pipermail/rails/attachments/20060514/93bb7173/attachment-0001.html

Hi. I am using BackgrounDRb to process thumbnails and upload to S3 - things are hunky-dory in development (thumbs are generated, these are uploaded to S3, the metadata is saved to trhe DB, and I get a nice status page updated by periodic calls via ask_status), but when I tried to deploy to our staging server and stop/start BackgrounDRb via Capistrano, things blew up - well, not exactly,

Hi all, I have Rails 4.0.0 app, After capistrano deploy my ckeditor_assets are being removed. Could not symlink to shared/ckeditor_assets folder. I have followed this step : namespace :uploads do > > desc <<-EOD > Creates the upload folders unless they exist > and sets the proper upload permissions. > EOD > task :setup,

I am new to RoR, but I have had RoR running with Capistrano on a different box. However I''m installing a fresh copy now and am finding constant regex errors from the standard.rb file while attempting to "show_tasks". As this is a new installation, everything is up-to-date. Here''s the environment: FreeBSD 6.2-RC2 ruby 1.8.5 gem 0.9.0 rake 0.7.1 net-ssh 1.0.10, 1.0.9

Does it make sense that if I type cap show_tasks it works but when I type rake show_tasks I get: rake aborted! Don''t know how to build task ''show_tasks'' /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake.rb:1287:in `[]'' /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake.rb:300:in `[]''

Hi everyone, I''ve been sold on the whole idea of using Capistrano for deployment of my sites and have successfully deployed three apps in three different environments all with success. As I''m trying to configure my sites I''ve encountered a number of issues/questions hence this post. 1. Database.yml Twice I''ve been stumped by apps refusing to work on

There seems to be a bug when deploying a rails application developed in windows to a linux box. The rake deploy command attempts to run the command "/install/path/current/script/process/reaper" but the files do not have the execute bit set. What can I do to make sure either the execute bit is run or prepend the command with sh or ruby? Thanks.

I have an application I''m working on which allows me to upload files. The files are stored somewhere in the public directory so that they can be accessed directly. When I deploy an update with Capistrano, I ''lose'' the files because of the current symlink points to the new checked out version. Before I start working on a solution, has anyone else already solved this

Hi, i''m striving to capistranize a foo app on DH. Everything is hosted on DH to simplify a bit : - svn repository (http access) - the application - the DB I used the recipe of Jamis Buck modified by Geoffrey Grosenbach which you can find here : http://nubyonrails.com/pages/shovel_dreamhost I did the least modifications possible. As recommended. 1) When I launch the

I have a few projects with large directories (say, user-uploaded files) that make using Capistrano a bit awkward. However, I suspect that I may have reached the "write a custom task" level. (I suppose I''m just thinking out loud, here...) It seems as though I''d want this directories symlinked in /shared... Maybe there''s a way to extend the

I am fighting with this for 5 days... i''m getting crazy! Here is my deploy.rb: require "bundler/capistrano" require "rvm/capistrano" # Load RVM''s capistrano plugin. set :rvm_ruby_string, :local # Or whatever env you want it to run in. ''1.9.3'' set :domain, ''www.xxx.com'' set :application,

My needs are far more modest than Capistrano''s capabilities. Also, when working in a shared hosting environment, the Capistrano model can be outside the capabilities allowed by the Web host. I simply need to push selected directories (on a changed-file basis) out to a server. I have about 80% of the code written to do this (I''m considering YAC for the name, with a nod to

Hello, I am trying to figure out how to use switch tower.I have a local copy of source code and I want to put it on the host.Now I dont want to put directories like config/ because the settings on the host and the one on my machine are different. Any way to do this? Also is svn necessary to use SwitchTower? Vivek -------------- next part -------------- An HTML attachment was scrubbed... URL:

Any idea why this command takes upwards of 15 minutes to complete? run "cd #{release_path} && ORACLE_HOME=/home/oracle RAILS_ENV=production rake db:migrate" When running it against my local Postgres it completes in a few seconds or so. Running it against my local Oracle XE takes only slightly longer. But running it with capistrano over ssh is taking years away from my life.

I had an incredibly difficult time getting this to work correctly and I have a couple of comments/questions, after the fact. When using caching and Capistrano together, how have you set the permissions on your RAILS_ROOT/public folder...or have you sent your cache data to your document root? I''ve got it working now, but of course, it will break the next time I deploy. I was

Hi all, I''m using Capistrano and SVN to version control and deploy, but I''m not sure what to do about my images directory. I use file_column and RMagick to save images to the disk under a model called "Image", and they''re saved in public/image/... etc. Each time I deploy a new version, my image folder doesn''t exist anymore. I don''t want