We're adding a script console-for to open a remote Rails console with one command. Also have a look at shell-for, which this script is relying on.

Run it from any project directory like this, passing a Capistrano multistage deployment target:

console-for staging

This script is part of our geordi gem on github.

When you have a hook in your Capistrano file that dumps your remote database, you might not want it to dump each time you deploy (say, you're experimenting with staging and don't want ten dumps an hour). How to skip dump creation:

Capistrano 2

In your Capistrano file:

before 'deploy:update_code', 'db:dump' unless fetch(:skip_dump, false)

The second parameter of fetch sets a default value if skip_dump is not defined.

In your terminal:

cap staging deploy -S skip_dump=true

The -S must be a capital letter t...

Soon after having written our shell-for script, we wanted to easily get dumps of our productions machines, too. This is how we do it:

dump-for staging [-s]

It will copy the dump to your project's tmp directory and name it according to the capistrano stage you're calling for, here: staging.dump. When you pass the optional -s option, the dump will automatically been sourced into your local development database.

This script ...

We regularly need to connect to the server in order to e.g. access the production console. Guessing the Capistrano deploy user and then again guessing the right directory on the server is awkward, so we wrote a script that parses config/deploy and gives you the handy command shell-for.

Run it from any project directory like this, passing a Capistrano multistage deployment target:

shell-for staging

Now it also supports commands to be remotely executed before loading the bash. Use --no-bash to only execute the command and load no ba...

We recently decided to put static content for HouseTrip.com to Amazon Cloudfront for a faster user experience. This happens fully automatically on deploy and is transparent in development. Together with a heavy use of sprites this sped up page load time quite nicely.

These are a couple of the problems you need to solve in order to do this:

• There is no good way to invalidate Cloudfront cached assets, and Cloudfront will ignor...

You can seriously speed up deployments with Capistrano when using a local git repository on the server you are deploying to.

Simply add

set :deploy_via, :remote_cache
set :copy_exclude, [ '.git' ]

to your config/deploy.rb and Capistrano will create a clone in shared/cached-copy. This will be updated using git pull when deploying which transfers less bytes and is usually much faster. If deploy_via is set to use default settings (being "export"), Capistrano will do a full clone of the repository from your git host otherwi...

After updating your RubyGems, you will probably not be able to run Capistrano any more, but receive an error similar to this:
can't activate net-ssh (= 2.0.22) for [], already activated net-ssh-2.1.0 for [] (Gem::LoadError)

If you have Bundler installed, you can use bundle exec to avoid this problem as follows:

Create a gemfile at ~/.capistrano/Gemfile (or at some other sensible place), that only contains these 2 lines:
source 'http://rubygems.org'
gem 'capistrano'
gem 'capistrano-ext' # You need this for multistag...

Note that this sounds good but is not good at all when hooking tasks on cap deploy (see this article). Make sure to hook your calls properly when using this.

To deploy an application and run all pending migrations before restarting it, you can use the following standard Capistrano task:

cap deploy:migrations

Little is known what happens when the deployment goes through, but a migration or a rake task aft...

In order to bring up a textual description of a Capistrano task you can say

cap -e taskname

... where taskname is the name of the task you're not sure about.

It's that simple to allow one of your Linux users to run a single command as UID 0:

1. sudo visudo
2. Add the line below to allow user 'deploy' to run /usr/bin/bundle with root privileges

deploy  ALL=NOPASSWD: /usr/bin/bundle

So you added a new callback to your model that (e.g.) caches some data when it is saved. Now you need to run that callback for the 10000 existing records in the production database. You have two options here:

1. Write a clever migration, possibly by embedding the model into the migration script.
2. Open the Rails console after deployment and re-save every single record. You should probably add two chores to [Pivotal Tracker](http://www.pivotaltrac...

You don't need a Rails application to use Sass. Even when you're working on a static site you can generate your CSS through Sass.

• Install Sass with sudo gem install haml
• Create a folder sass in the folder, that stores your stylesheets, e.g. mkdir css/sass
• In a separate terminal window, run sass --watch css/sass:css. This will watch your sass files for changes and rewrite stylesheets as required.

This even works on Windows.

Note about your .gitignore

You might want to change our [typical .gitignor...

When Paperclip attachments should only be downloadable for selected users, there are three ways to go.
The same applies to files in Carrierwave.

1. Deliver attachments through Rails

The first way is to store Paperclip attachments not in the default public/system, but in a private path like storage inside the current release. You should prefer this method when dealing with sensitive data.

Make ...

Update RubyGems and Passenger

Bundler requires Rubygems >= 1.3.6. Run gem update --system if you have an older version.
It also is not compatible with older versions of passenger, so bring that up to date as well (2.2.15 works).

If you installed RubyGems through apt (which you should never do!), you may see a message giving you a hint to use apt to update.
Some people advise to install the 'rubygems-update-1.3.7' gem on Ubuntu systems if you used apt to install RubyGems.
I did that - and lost all...

So you screwed up and copied Paperclip secrets from one project to another. Here is a semi-automatic, painful way to migrate your existing attachment files to new locations.

You need to follow this step by step, do not just copy the whole thing into the console!

# 1. Get old paths by doing something like this on the console:
old_paths = ModelWithAttachment.all.collect { |m| [m.id, File.dirname(m.image.path(:original)).gsub(/original$/, '') ] if m.image.file? }.compact.uniq

# 2. Now change the Paperclip secret on the co...

You are getting when connecting via SSH or deploying with Capistrano (which uses SSH):

Too many authentication failures for username

This is caused by having too many SSH keys added to your keyring or ssh-agent. Your ssh-agent will throw all keys against a server until one matches. Most servers will deny access after 5 attempts.

This issue might come and go as the order of the active SSH keys in your ssh-agent changes.

Quick fix

Have less keys. Up to 5 keys are fine when the SSHD you're connecting to is using the default ...