Find out when a scheduled job didn't run at all
The worst failures leave no trace: the scheduled task or cron job simply didn't run. There is no error, no exit code and no log, so nothing you have can alert you. The usual reasons are few, and the fix is the same for all of them: expect the run, and alert when it's missing.
Why scheduled jobs silently skip runs
On Windows, Task Scheduler has several settings that quietly prevent a run, most of them on by default or easy to miss:
- The computer was off or asleep. A task doesn't wake the PC unless Wake the computer to run this task is ticked (and the power plan allows wake timers). If the start time passes, the run is skipped, unless Run task as soon as possible after a scheduled start is missed is ticked, which is off by default.
- On battery. Start the task only if the computer is on AC power is on by default for every task, and since Windows 10 the battery saver delays most triggers as well.
- Nobody is logged on. A task set to Run only when user is logged on doesn't run on a server or PC where you are logged out.
- A repetition that depends on one start. "Repeat every hour" hangs off one daily trigger; if that start is missed, the hourly runs can be missing for the whole day (a case discussed on Microsoft Q&A in 2025).
On Linux, cron doesn't catch up either: a job whose minute passes while the machine is off is simply not run (anacron exists for this, but works in days, not hours). A broken crontab line, a stopped cron service or a moved script give the same silence.
Any check that waits for something to happen — an error email, an event-log trigger, the job's own log — has nothing to react to. The only reliable way is the other direction: something outside the machine knows when a run is due and complains when it doesn't hear from the job.
Expect a ping, alert when it doesn't come
In TaskBeacon each job has a schedule, like the trigger in Task Scheduler: daily at 06:00, on weekdays, on the 1st of the month, every 2 hours. The job sends one ping when it finishes. If a planned run plus a grace period passes without that ping, the job turns Missed and you get an alert. The smallest version is just the last line, here in Python:
# --- TaskBeacon: top of the script (standard library only) ---
import json, urllib.request
TB = "https://taskbeacon.io/ping/your-job-id"
def tb_ping(path="", data=None):
body = json.dumps(data).encode() if data else None
req = urllib.request.Request(TB + path, data=body, headers={"Content-Type": "application/json"})
try:
urllib.request.urlopen(req, timeout=10)
except Exception as e:
print("TaskBeacon ping failed:", e)
# --- your code here ---
df = [1, 2, 3] # example data: replace with your own (e.g. pandas.read_csv)
# --- TaskBeacon: end of the script ---
tb_ping()In the app, the Code tab of your job gives you this code with the job's own address instead of your-job-id, and you can untick the parts you don't need.
Any language works the same way, with the code from its Docs page; for a program you can't change, use the PowerShell wrapper or a cron line. Adding the start ping and the failure ping (the default code) also tells you about runs that started and never finished, and about errors.
Getting the timing right
- Grace is how late a run may report before it counts as Missed. After a few runs TaskBeacon suggests a grace that would have covered your real runs. See Grace.
- No-alert days cover holidays: nothing is planned and nothing counts as Missed.
- A whole machine: when every job on one PC is missed at once, you get one alert for the machine instead of one per job, which usually means the PC is off or lost its network.
- A new job stays silent until its first ping, so setting it up never sends a false Missed.
Without TaskBeacon
Write a "watchdog" job that runs later in the morning and checks that each job's output file or log is from today, and emails you if not. On Windows, also tick Run task as soon as possible after a scheduled start is missed and untick the AC-power condition; on Linux, consider anacron for daily jobs. The catch is that the watchdog usually runs on the same machine, so when that machine is off, the watchdog is off too. A check that runs elsewhere — a heartbeat — doesn't have that blind spot.
Connected Claude or ChatGPT to TaskBeacon? Ask it "what failed last night, and why?" and it answers from your jobs. How that works.
Related
In the Docs: Schedules · Statuses · A whole machine goes quiet.
- Heartbeat monitoring (a dead man's switch) for scheduled scriptsThe job checks in on every run; silence is the alert.
- Task Scheduler can't send email any more: get an alert when a task failsThe built-in e-mail action is gone; event-triggered tasks miss runs that never happened.
- Monitor an Excel VBA macro run by Task SchedulerUnattended Excel tends to hang on a dialog rather than fail.
TaskBeacon is free to start, with every feature open for now and no credit card. Create a job, paste the code, and the first ping shows up on the job page.
Start free View live demoSources
- Microsoft Learn: TaskSettings.StartWhenAvailable (run after a missed start; off by default) (2018, still current; checked 2026-11-09)
- Microsoft Learn: TaskSettings.DisallowStartIfOnBatteries (true by default) (2018, still current; checked 2026-11-09)
- Microsoft Learn: What's New in Task Scheduler (battery saver delays most triggers) (2018-05-31, updated 2025-03-11; still the current page)
- Microsoft Q&A: Scheduled task not waking computer (2025-12-12)
- Microsoft Q&A: Task Scheduler doesn't run the task (hourly repetition skipped) (2025-08-07)
- man7.org: anacron(8) (runs jobs that were missed while the machine was off, in days) (page 2012, matches upstream 2026)
Page checked against its sources on 2026-11-09. Menus and limits of other products change; if something here no longer matches, write to support@taskbeacon.io.