[openstreetmap/openstreetmap-website] Implement Delayed Job as an Active Job backend (#2015)

[Andy Allan]

In #1852 we're looking at processing GPX files using the [Active Job](https://edgeguides.rubyonrails.org/active_job_basics.html) framework, and there's many other places where it makes sense to start using job queues too (for example, sending multiple diary entry comment notifications) for tasks that could be run asynchronously.

ActiveJob comes with a simple built-in task runner, but that uses the current process and so isn't really production-grade. I've had good experiences with [Delayed Job](https://github.com/collectiveidea/delayed_job), which stores jobs in PostgreSQL. The alternative would be e.g. resque using redis for job persistence, but I don't think that's necessary since it's something else to run, and we're not exactly creating bazillions of jobs per second.

Since I'm fairly confident that we'll use Active Job for multiple topics, I think it's worth setting up the backend (migrations, configuration and deploying task runners in production) separately, to avoid having to combine all this stuff into a larger PR.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/2015
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/rails-dev/attachments/20181005/8f9e6be5/attachment.html>