pshima

namespace :assets do
  task :generate, :except => { :no_release => true }, :roles => :app do
    desc "Asset copy"
    run "mkdir #{release_path}/public/oldassets"
    run "mkdir #{release_path}/public/oldassets/stylesheets"
    run "mkdir #{release_path}/public/oldassets/javascripts"
    run "cp -r #{deploy_to}/current/public/stylesheets/* #{release_path}/public/oldassets/stylesheets/"
    run "cp -r #{deploy_to}/current/public/javascripts/* #{release_path}/public/oldassets/javascripts/"
  end
end

pshima

    location ~ ^/(images|assets|javascripts|stylesheets|fonts)/ {
      try_files /$uri /oldassets/$uri =404;
      expires 10y;
    }

pshima

command[check_unicorn]=/usr/lib/nagios/plugins/check_procs -c 1:24 -C unicorn

pshima

command[check_raindrops]=/usr/lib/nagios/plugins/check_raindrops -w 512 -c 900

pshima

# from https://github.com/petey5king/raindrops-nagios

#!/usr/bin/env ruby

#script to check the unicorn backlog

#requires the nagios gem, raindrops and rubygems

$LOAD_PATH.unshift '../lib'
require 'rubygems'

pshima

 before_migrate do
      if app['main']['uses_bundler']
        link "#{release_path}/vendor/bundle" do
          to "#{app['main']['deploy_to']}/shared/vendor_bundle"
        end
        common_groups = %w{development test cucumber staging production}
        execute "LANG=en_US.UTF-8 LC_ALL='en_US.UTF-8' bundle install --deployment --without #{(common_groups -([app['main']['rails_env']])).join(' ')}" do
          ignore_failure true
          cwd release_path
          environment ({'LANG' => 'en_US.UTF-8'})

pshima

    redis = Redis.new(:host => config[:host], :port =>config[:port])

    redis.info.each do |k, v|
      output "#{config[:scheme]}.#{k}", v
    end

pshima

  haproxy_http_production:       
    command:      /etc/sensu/plugins/haproxy/haproxy-metrics.rb -h my.hostname.com -q haproxy_stats -u username -p password -s my.hostname
    handlers:     ["graphite"]
    interval:     60
    subscribers:  ["sensu_server"]
    type:         metric

  memcached_graphite_production: 
    command:      /etc/sensu/plugins/memcached/check-memcached-graphite.rb -h my.hostname -p 11211
    handlers:     ["graphite"]

pshima

/path/to/unicorn/log/unicorn.stderr.log
/path/to/production/log/production.log 
{
  daily
  missingok
  rotate 180
  compress
  dateext

  # this is important if using "compress" since we need to call

pshima

Runbook for X

What is happening?

Such and such service is probably having X, that means Y.

What is the user experience when this happens?

No users will be able to do anything.