Delayed Job

What Delayed Job is

Delayed Job is a self-hosted, database-backed priority queue for running longer tasks asynchronously in the background. Listed uses include newsletter sending, image resizing, HTTP downloads, batch imports, and spam checks. Methods can be queued with `.delay`, with settings for priority, scheduled run time, and named queues. Separate worker processes can be assigned to selected queues or worker pools. Rails 4.2 and later can use Delayed Job as the Active Job queue adapter, and the documentation describes Active Record and Mongoid storage backends. Failed jobs are rescheduled with delays based on their attempt count, up to 25 attempts by default. After retries are exhausted, jobs are deleted by default, although a worker can be configured to keep them in the database. The default maximum runtime is four hours; a longer-running job could be picked up by another computer. The current gem specification requires Ruby 3.2 or later and ActiveSupport 3.0 or later but below 9.0. The software is free under the MIT license, which provides it as-is without warranty.

Who it suits

Delayed Job suits Ruby and Rails developers who need a self-hosted queue for background tasks. It is especially relevant to Rails 4.2-or-later projects using Active Job and teams able to meet its Ruby and ActiveSupport requirements.

Limits

The default maximum job runtime is four hours, and the default maximum is 25 retry attempts. Failed jobs are deleted after retries unless workers are configured to keep them in the database.

Getting started

1. Open the Delayed Job project page on GitHub.
2. Check that your environment meets the Ruby and ActiveSupport requirements.
3. Choose Active Record or Mongoid for queue storage.
4. For Rails 4.2 and later, configure Delayed Job as the Active Job queue adapter.
5. Queue methods with `.delay` and set any needed priority, schedule, or named queue.
6. Start background worker processes and assign them to queues or worker pools.

Verdict

Delayed Job provides queueing, scheduling, and retry controls for background work in a self-hosted package. Check the runtime defaults and Ruby and ActiveSupport requirements, and plan how failed jobs should be retained.