How to Prevent Pilled Up Job Executions in GoCron if Job Takes Longer than Scheduled Interval?

[iyashjayesh]

I'm using the github.com/go-co-op/gocron package in my Go project to schedule tasks. The job is set to run every 2 minutes, but there's a scenario where the task might take longer than the interval (e.g., the job takes 7 minutes to complete). After the long-running job completes, multiple job instances get triggered back-to-back, which causes overlapping executions.

func newScheduler(loc *time.Location) {
    e := gocron.NewScheduler(loc)
    e.SingletonModeAll()

    e.Every("2m").Do(func() {
        Logger.Warn("Task Started at: " + time.Now().Format("2006-01-02 15:04:05"))
        err := tasks.MyTask()
        if err != nil {
            Logger.Error("Error occurred during task execution: ", err)
        }
        Logger.Warn("Task Completed at: " + time.Now().Format("2006-01-02 15:04:05"))
    })

    e.StartAsync()
    select {}
}

Problem:
When the job takes more time than the scheduled interval (e.g., 7 minutes instead of 2), multiple instances of the job get executed back-to-back after the long-running job finishes. This causes performance issues and overlapping jobs.

Question:
Is there a way to ensure that only one instance of the job runs after execution of the job, and subsequent jobs wait until the previous instance finishes before starting and should only once based on the next interval?

What I've tried:
I've enabled SingletonModeAll(), but it doesn't prevent the execution of the back-to-back jobs when the interval is missed due to a long-running task.

Desired Outcome:
I want to ensure that after execution of the job, it should ignore the pilled-up job and should not run back to back, even if the job takes longer than the scheduled interval.

Any guidance on how to handle this in GoCron or alternative approaches would be greatly appreciated!

[JohnRoesler]

Looks like you're using v1. If you upgrade to V2, you can use Reschedule with singleton mode https://pkg.go.dev/github.com/go-co-op/gocron/v2#example-WithSingletonMode

[iyashjayesh]

Yeah, this will work. Thanks @JohnRoesler 😄

[iyashjayesh]

@JohnRoesler, the function you provided helped me with the accumulated issue. However, when I have multiple jobs, I need to wait for the timer to run the jobs.

Is there a way to execute the jobs immediately when the scheduler starts, and then have the scheduler continue running at the specified interval?

I tried adding the bellow param in NewJob and it's working, but is this the right away or do we have any other to achieve this?

gocronv2.WithStartAt(
	gocronv2.WithStartImmediately(),
),

Example JYFR,

func startScheduler() {
	log.Println("Starting Scheduler")
	s, _ := gocronv2.NewScheduler()
	defer func() { _ = s.Shutdown() }()

	log.Println("Job 1")
	s.NewJob(
		gocronv2.DurationJob(
			2*time.Minute,
		),
		gocronv2.NewTask(
			func() {
				log.Println("Function 1 Started")
				randomTime()
				log.Println("Function 1 Ended")
			},
		),
		gocronv2.WithSingletonMode(1),
	)

	log.Println("Job 2")
	s.NewJob(
		gocronv2.DurationJob(
			2*time.Minute,
		),
		gocronv2.NewTask(
			func() {
				log.Println("Function 2 Started")
				randomTime()
				log.Println("Function 2 Ended")
			},
		),
		gocronv2.WithSingletonMode(1),
	)

	s.Start()
	select {}
}

[JohnRoesler]

That's the right way