Framework agnostic Rufus-scheduler wrapper to run recurring jobs.
Add this line to your application's Gemfile:
gem "schked"And then execute:
bundleOr install it yourself as:
gem install schkedSchked requires Ruby 2.7+.
The test matrix covers Ruby 2.7, 3.0, 3.1, 3.2, 3.3, 3.4, and 4.0. Rails integration tests run on every Ruby; Rails 8 is only included on Ruby 3.2+.
.schked
--require config/environment.rb
config/schedule.rb
cron "*/30 * * * *", as: "CleanOrphanAttachmentsJob", timeout: "60s", overlap: false do
CleanOrphanAttachmentsJob.perform_later
endIf you have a Rails engine with own schedule:
engine-path/lib/foo/engine.rb
module Foo
class Engine < ::Rails::Engine
initializer "foo" do |app|
Schked.config.paths << root.join("config", "schedule.rb")
end
end
endAnd run Schked:
bundle exec schked startTo show schedule:
bundle exec schked showWhen you deploy your schedule to production, you want to start new instance before you shut down the current. And you don't want simultaneous working of both. To achieve a seamless transition, Schked is using Redis for locks.
You can configure Redis client as the following:
Schked.config.redis = {url: ENV.fetch("REDIS_URL") }Also, you can define callbacks for errors handling:
config/initializers/schked.rb
Schked.config.register_callback(:on_error) do |job, error|
Raven.capture_exception(error) if defined?(Raven)
endThere are :before_start, :after_finish and :around_job callbacks as well.
Warning: :before_start and :after_finish callbacks are executed in the scheduler thread, not in the work threads (the threads where the job execution really happens).
:around_job callback is executed in the job's thread.
Schked.config.register_callback(:around_job) do |job, &block|
...
block.call
...
endBy default Schked writes logs into stdout. In Rails environment Schked is using application logger. You can change it like this:
config/initializers/schked.rb
Schked.config.logger = Logger.new(Rails.root.join("log", "schked.log"))Schked can expose a small HTTP endpoint for Kubernetes liveness probes. It is disabled by default to keep the existing behavior unchanged.
Configure it in Ruby:
Schked.config.liveness_probe = {
enabled: true,
bind: "0.0.0.0",
port: 8080,
path: "/healthz",
heartbeat_interval: 5,
heartbeat_threshold: 15
}Or via CLI flags:
bundle exec schked start --liveness-probe --liveness-bind 0.0.0.0 --liveness-port 8080 --liveness-path /healthzIn Rails, set it through the application config:
# config/application.rb or config/environments/*.rb
config.schked.liveness_probe = {
enabled: true,
bind: "0.0.0.0",
port: 8080,
path: "/healthz",
heartbeat_interval: 5,
heartbeat_threshold: 15
}The endpoint returns 200 OK while the scheduler is responsive and 503 Service Unavailable when the heartbeat is stale or during shutdown. The scheduler updates the heartbeat every heartbeat_interval seconds (default 5); if it is not updated within heartbeat_threshold seconds (default 15), the endpoint reports unhealthy. Use it in Kubernetes like this:
livenessProbe:
httpGet:
path: /healthz
port: 8080
initialDelaySeconds: 10
periodSeconds: 10Yabeda::Schked - built-in metrics for monitoring Schked recurring jobs out of the box! Part of the yabeda suite.
describe Schked do
let(:worker) { described_class.worker.tap(&:pause) }
around do |ex|
Time.use_zone("UTC") { Timecop.travel(start_time, &ex) }
end
describe "CleanOrphanAttachmentsJob" do
let(:start_time) { Time.zone.local(2008, 9, 1, 10, 42, 21) }
let(:job) { worker.job("CleanOrphanAttachmentsJob") }
specify do
expect(job.next_time.to_local_time)
.to eq Time.zone.local(2008, 9, 1, 11, 0, 0)
end
it "enqueues job" do
expect { job.call(false) }
.to have_enqueued_job(CleanOrphanAttachmentsJob)
end
end
endBug reports and pull requests are welcome on GitHub at https://github.com/bibendi/schked. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the Contributor Covenant code of conduct.
The gem is available as open source under the terms of the MIT License.
Everyone interacting in the Schked project’s codebases, issue trackers, chat rooms and mailing lists is expected to follow the code of conduct.