Advanced usage

This section documents advanced usage patterns.

Guide map

• With Celery — keep schedium for planning and dispatch due work to Celery workers.

• Timezones — recommendations for UTC-aware scheduling and DST caveats.

• Handling exceptions in jobs — failure behavior, retry semantics, and cancellation helpers.

• Threading schedulers — run jobs concurrently with thread pools/queues.

When to use this section

Use these guides if you need one or more of the following:

• integration with external execution systems,

• predictable behavior across timezones and DST,

• explicit error-handling strategies,

• concurrent execution while preserving trigger semantics.

• With Celery
  • Example: dispatch Celery tasks when due
  • Notes
• Timezone
  • Preferred: UTC-aware datetimes
  • Using local time
  • Naive datetimes
• Exceptions
  • What happens by default
  • Patterns
  • Cancel a job after a failure (decorator)
  • Exceptions from next-run computations
• Threading
  • When to use which
  • ThreadedJobsScheduler (thread pool)
  • SchedulerThread (scheduler loop in background thread)
  • QueuedJobsScheduler (producer queue + worker threads)
  • Notes and caveats