QueueTimeStatsdReporter

QueueTimeStatsdReporter reports to Datadog the time a request spends queued before hitting your Rails/Rack application.

Add this line to your application's Gemfile:

gem 'queue_time_statsd_reporter'

And then execute:

$ bundle

Or install it yourself as:

$ gem install queue_time_statsd_reporter

In your web server, configure the X-Request-Start header, eg. for Nginx:

location @app {
  proxy_set_header X-Request-Start "t=${msec}";
}

Next in Rails create an initialiser containing the code:

metric_name = "mycorp.widgetsapp.queue_time"
tags = ["env:#{Rails.env}"]
Rails.application.middleware.unshift QueueTimeStatsdReporter, my_statsd_object, metric_name, tags

where my_statsd_object is an object, perhaps a proxy for your statsd implementation, that has a method with signature:

gauge(metric, value, tags=[])

When a request is received QueueTimeStatsdReporter will report the time spent queueing by calling the guage method with:

• metric: The string metric name, eg. "mycorp.widgetsapp.queue_time"
• value: the time spent queued in seconds
• tags: any tags passed into the middleware initialiser

For instance, if you're reporting to Datadog your object might look something like:

class MyDatadogClass
  def initialize(endpoint = Datadog::Statsd.new(ENV.fetch("STATS_HOST", "localhost"), 8125))
    @endpoint = endpoint
  end

  def gauge(metric, value, tags=[], sample_rate=1)
    @endpoint.gauge(metric, value, tags: tags, sample_rate: sample_rate)
  end
end

and your initialiser:

Rails.application.middleware.unshift QueueTimeStatsdReporter, MyDatadogClass.new, "mycorp.widgetsapp.queue_time", ["env:#{Rails.env}", "infra:aws"]

You can use this middleware outside of Rails too.

After checking out the repo, run bin/setup to install dependencies. Then, run rake spec to run the tests. You can also run bin/console for an interactive prompt that will allow you to experiment.

To install this gem onto your local machine, run bundle exec rake install. To release a new version, update the version number in version.rb, and then run bundle exec rake release, which will create a git tag for the version, push git commits and tags, and push the .gem file to rubygems.org.

Bug reports and pull requests are welcome on GitHub at https://github.com/rubytune/queue_time_statsd_reporter.

The gem is available as open source under the terms of the MIT License.

• @wjessop, will@willj.net