How to Monitor Cron Jobs and Scheduled Python Scripts

Python powers a huge share of scheduled work — ETL pipelines, scraping, ML retraining, report exports. However you schedule it, the failure mode is the same: when it stops, it stops quietly. A heartbeat ping fixes that.

Plain cron + curl

The simplest approach pings after the script exits successfully:

0 * * * * /usr/bin/python3 /app/etl.py && curl -fsS https://api.cronguard.dev/v1/ping/<your-uuid>

Pinging from inside Python

For start/success/failure signalling, ping directly from the script. Using requests:

import requests

BASE = "https://api.cronguard.dev/v1/ping/<your-uuid>"

def ping(suffix=""):
    try:
        requests.get(BASE + suffix, timeout=10)
    except requests.RequestException:
        pass  # never let monitoring break the job

def main():
    ping("/start")
    try:
        run_etl()
        ping()          # success
    except Exception:
        ping("/fail")
        raise

if __name__ == "__main__":
    main()

APScheduler

Wrap the job, or add a listener that pings on success and failure events:

from apscheduler.events import EVENT_JOB_EXECUTED, EVENT_JOB_ERROR
import requests

BASE = "https://api.cronguard.dev/v1/ping/<your-uuid>"

def on_event(event):
    suffix = "/fail" if event.exception else ""
    try:
        requests.get(BASE + suffix, timeout=10)
    except requests.RequestException:
        pass

scheduler.add_listener(on_event, EVENT_JOB_EXECUTED | EVENT_JOB_ERROR)

Celery beat

For a periodic Celery task, ping at the end of the task body — or use task_success / task_failure signals to ping centrally for all periodic tasks.

Alerts that reach you

Send those pings to CronGuard and define how often each job should check in. If a Python job misses its window — crash, hang, broken schedule or dead host — CronGuard alerts you instantly across multiple channels and tracks how long each run takes. Free to start, no credit card.