SchedulerStrategy
SchedulerStrategy
This strategy is used to define the mechanism by which scheduled tasks are executed and how they are reported on. The main purpose of this strategy is to ensure that a given task is executed exactly once at the scheduled time, even if there are multiple instances of the worker running.
To do this, the strategy must use some form of shared storage and a method of locking so that only a single worker is allowed to execute the task.
By default, the DefaultSchedulerStrategy will use the database to enable this functionality.
interface SchedulerStrategy extends InjectableStrategy {
executeTask(task: ScheduledTask): (job: Cron) => Promise<any> | any;
getTasks(): Promise<TaskReport[]>;
getTask(id: string): Promise<TaskReport | undefined>;
updateTask(input: UpdateScheduledTaskInput): Promise<TaskReport>;
}
- Extends:
InjectableStrategy
executeTask
(task: ScheduledTask) => (job: Cron) => Promise<any> | anyExecute a scheduled task. This method must also take care of ensuring that the task is executed exactly once at the scheduled time, even if there are multiple instances of the worker running.
For instance, in the DefaultSchedulerStrategy we make use of a dedicated database table and a locking mechansim. If you implement a custom SchedulerStrategy, you must use some other form of shared locking mechanism that could make use of something like Redis etc. to ensure that the task is executed exactly once at the scheduled time.
getTasks
() => Promise<TaskReport[]>Get all scheduled tasks.
getTask
(id: string) => Promise<TaskReport | undefined>Get a single scheduled task by its id.
updateTask
(input: UpdateScheduledTaskInput) => Promise<TaskReport>Update a scheduled task.
TaskReport
A report on the status of a scheduled task.
interface TaskReport {
id: string;
lastExecutedAt: Date | null;
isRunning: boolean;
lastResult: any;
enabled: boolean;
}