将rufus-scheduler调度到heroku的dyno上

Question

I'd like to use rufus-scheduler to have a background task run. I currently have a worker dyno for Sidekiq and would ideally just like to schedule onto that process. Testing the rufus-scheduler as per the documentation at https://github.com/jmettraux/rufus-scheduler#so-rails

require 'rufus-scheduler'

# Let's use the rufus-scheduler singleton
#
s = Rufus::Scheduler.singleton


# Stupid recurrent task...
#
s.every '1s' do
  Rails.logger.info "hello, it's #{Time.now}"
end

It comes out on both the web and worker and I'd like it to use the worker process. How would I achieve that?

Sample output is:

2014-10-17T00:30:29.382689+00:00 app[web.1]: hello, it's 2014-10-17 00:30:29 +0000
2014-10-17T00:30:29.609192+00:00 app[worker.1]: hello, it's 2014-10-17 00:30:29 +0000

Answer 1

Something like the following could do it:

if running_on_worker_process? # you have to implement that one yourself.
  require 'rufus-scheduler'
  s = Rufus::Scheduler.singleton
  s.every '1d' do
    Rails.logger.info "taking out the garbage..."
  end
end

Beware your worker process (dyno?) going to sleep (and the scheduler with it).

EDIT 1

Tip: you can use the $DYNO variable ( https://devcenter.heroku.com/articles/dynos#local-environment-variables ) to identify which dyno you're on.

if ENV['DYNO'].match(/^worker\./)
  # ...
end