Documentation: Where and when should JobScheduler.schedule() be called?

[henriquealbert]

The documentation shows how to use JobScheduler.schedule() but doesn't explain where to call it or which AdonisJS environment to use.

Should scheduled jobs be set up in a preload file? If so, which environment?

Current Approach (uncertain if correct)

// start/scheduler.ts
 import app from '@adonisjs/core/services/app'
 import { JobScheduler } from '@nemoventures/adonis-jobs'
 import MyJob from '#jobs/my_job'

 app.ready(async () => {
   await JobScheduler.schedule({
     key: 'my-scheduled-job',
     job: MyJob,
     data: {},
     repeat: { pattern: '0 * * * *' },
   })
 })

  // adonisrc.ts
  preloads: [
    {
      file: () => import('#start/scheduler'),
      environment: ['console'], // ← Is this correct?
    },
  ]

What needs clarification

1. Is using a preload file the recommended approach?
2. Should environment: ['console'] be used, or no environment restriction?
3. Is JobScheduler.schedule() idempotent (safe to call multiple times)?
4. In multi-process deployments (separate web/worker containers), where should schedules be registered?

Suggested addition

A "Scheduling Jobs" guide section covering the recommended setup pattern and environment configuration would be very helpful.

[Julien-R44]

JobScheduler can be called anywhere you want. Personally, we use it inside an AdonisJS provider, in the ready method.

Basically, what JobScheduler does is just store data in Redis, that will be picked up later by the workers ( node ace queue:work ) . So you can really call it wherever you prefer, in a provider, a preload file, or any environment. You can even call it from a specific endpoint; for example, imagine an endpoint that, once triggered, needs to run a task every hour. In that case, using JobScheduler inside your controller makes perfect sense and works perfectly fine

Let me know if that makes sense, and yeah, maybe we could update the docs to reflect that

[henriquealbert]

Yeah that makes sense! Thanks!