Task Scheduling

% warning: This feature is only accessible by direct manipulation of % TaskSchedule objects. It is targeted for plugin writers and no api access is planned.

Pulp supports scheduling of tasks. Scheduled tasks will be dispatched shortly after their next_dispatch time, and be rescheduled one dispatch_interval after that, if the latter is set. By specifying the dispatch_interval as time_delta(days=1) you can expect the task dispatch to stabily happen at same time every day. Until the last task of the same schedule enters a final state, a new task will not be dispatched. Scheduling is done by the worker processes, therefore scheduled task dispatching will be missed if all workers are offline. After an outage window, overdue schedules will dispatch at most one task, but down to timing, they may be rescheduled shortly thereafter. The task schedule API at /pulp/api/v3/task-schedules/ is provided to read the tasks schedules.

Passing Arguments to Scheduled Tasks

The task_args and task_kwargs fields on TaskSchedule allow plugin writers to store positional and keyword arguments that will be forwarded to the task function each time it is dispatched. This is useful when a scheduled task needs to operate on a specific resource or with specific options.

from datetime import timedelta
from pulpcore.plugin.models import TaskSchedule

TaskSchedule(
    name="my-plugin-sync-schedule",
    task_name="my_plugin.app.tasks.sync",
    task_kwargs={"remote_pk": str(remote.pk), "optimize": True},
    dispatch_interval=timedelta(hours=6),
).save()

The args and kwargs stored in task_args and task_kwargs are passed directly to dispatch() when the schedule fires, so they should match the signature of the task function referenced by task_name. Both fields are encrypted at rest (they will not be included in API responses).