Create a monitor
Choose the job’s expected interval, grace period, and maximum runtime. PulseWatch gives you a private ping URL.
Monitoring for unattended work
Know when your scripts stop running – before your users do.
Logs can tell you when a job fails. They cannot tell you when it never ran. PulseWatch watches cron jobs, scripts, and automations from the outside, then alerts you when they go missing, crash, or get stuck.
Free for 2 monitors · No credit card · No SDK
External watchdog active
UTC
NIGHTLY JOB
Database backup
MISSING
No successful run within the expected window.
Last successExpectedAlert sent
The useful failure signal is the one your script cannot send itself.
PulseWatch checks independently.
How it works
A few minutes to set up. Quiet until it matters.
Keep your existing scheduler and code. Add a lightweight start and outcome signal.
Add lightweight pings
Call the URL when work starts, succeeds, or fails. It works from any language that can make an HTTP request.
Let the watchdog notice absence
The independent watchdog evaluates monitors every 5–30 minutes, depending on plan, and alerts on state changes.
No monitoring dependency
Use Python’s standard library.
A short timeout and a guarded ping keep a monitoring outage from interrupting the job being observed.
Read the complete integration guidefrom urllib.request import urlopen
PING_URL = "https://pulsewatch.example/ping/example-token"
def ping(outcome):
try:
urlopen(f"{PING_URL}/{outcome}", timeout=5).close()
except Exception:
pass
ping("start")
try:
your_existing_job()
except Exception:
ping("fail")
raise
else:
ping("success")Built for solo operators
Operational clarity without an observability project.
Made for unattended jobs
Cron jobs, scheduled scripts, scrapers, data pipelines, and AI workflows—anything that should run without you watching.
Transition-based alerts
One notification when a state changes, one when it recovers. Ordinary alerts are suppressed while a monitor is flapping.
A real dead-man’s switch
The separate watchdog catches the failure no in-process logger can: a job that never started.
The product
Evidence first when something goes wrong.
Current health, failure context, and recent runs stay ahead of configuration.
Simple pricing
Start free. Upgrade when you need more.
Free
$0 / month
Try it out, no card needed.
- 2 monitors
- Email alerts
- 7-day run history
- 30-minute checks
Solo
$7 / month
For developers running several automations.
- 10 monitors
- Email alerts
- Unlimited run history
- 10-minute checks
Indie
$19 / month
For power users with many pipelines.
- Unlimited monitors
- Email alerts
- Unlimited run history
- 5-minute checks
- Priority support