A tool for managing your environment variables across your cluster. The library
manipulates the system-wide environment variables by modifying the
/etc/environment
file.
cap env:list
– To list current environment configurationcap env:get[VARIABLE_NAME, VARIABLE_NAME, ...]
– To get the value of a list of variablescap env:unset[VARIABLE_NAME, VARIABLE_NAME, ...]
– To delete environment variablescap env:set[VARIABLE_NAME=VALUE, VARIABLE_NAME=VALUE, ...]
– To set environment variablescap env:sync
– To synchronise the environment configuration across servers
If you need to programatically manipulate the environment you can use the
Capistrano::EnvConfig::Environment
class like so:
require 'capistrano/env_config/environment'
environment = Capistrano::EnvConfig::Environment.new # Reads and loads /etc/environment from all servers
environment.list # Returns a hash of all environment variables
environment.get( 'VARIABLE' ) # Gets a variable by name
environment.set( 'VARIABLE', 'VALUE' ) # Sets a variable
environment.delete( 'VARIABLE' ) # Deletes a variable
environment.sync # Uploads current confifuration to all servers
By default capistrano sets and reads the /etc/environment
of all roles. This
behaviour can be changed by setting:
set :env_config_roles, [ :web, :app ]
The new
and sync
methods also accept an optional argument for specifying
roles explicitly (overriding the value set in :env_config_roles
):
environment.list( :app )
environment.sync( [ :web, :app ] )
Add the gem to your Gemfile
:
group :development do
gem 'capistrano-env_config'
end
And the following line to your Capfile
:
require 'capistrano/env_config'
In your server provisioning script, make sure the deploy user has write access
to the /etc/environment
file, which is used to store the configured variables.
The simplest way to do that would be to:
chown :deploy /etc/environment
chmod g+w /etc/environment
Copyright Itay Grudev (c) 2017.
This gem and all associated documentation is distributed under the terms of the MIT license.