bradens / unicorn-prewarm

Prewarm your Unicorn workers before real requests come in

Unicorn::Prewarm []

Prewarm your workers before they receive real traffic.

Installation

Add this line to your application's Gemfile:

gem 'unicorn-prewarm'

And then execute:

$ bundle

Or install it yourself as:

$ gem install unicorn-prewarm

Usage

In unicorn config after_fork do:

after_fork do |server,worker|
  response = Unicorn.prewarm(server)
  # check if the response is OK, it shall be Net::HTTPSuccess
end

Rack::Timeout

If you are using Rack::Timeout you might want to do something like:

after_fork do |server, worker|
  # Code to reconnect all databases belongs here
  server.logger.info("worker=#{worker.nr} spawned pid=#{$$}")

  timeout = Rack::Timeout.service_timeout
  begin
    Rack::Timeout.service_timeout = 60
    server.logger.info("worker=#{worker.nr} prewarming")
    start = Time.now
    Unicorn.prewarm(server).is_a?(Net::HTTPSuccess) or raise "Prewarm of worker #{worker.nr} failed"
    server.logger.info("worker=#{worker.nr} prewarmed in #{Time.now - start}")
  rescue => error
    server.logger.error("worker=#{worker.nr} prewarm failed: #{error} in #{Time.now - start}")
    raise
  ensure
    Rack::Timeout.service_timeout = timeout
  end
end

To prevent from timeouting on first requests.

Contributing

1. Fork it ( https://github.com/3scale/unicorn-prewarm/fork )
2. Create your feature branch (git checkout -b my-new-feature)
3. Commit your changes (git commit -am 'Add some feature')
4. Push to the branch (git push origin my-new-feature)
5. Create a new Pull Request