Skip to content

Know when a cron job stays quiet.

A heartbeat monitor gives each job a secret URL. The job calls it when it finishes. Silence, a failed run, or a run that takes too long is what UPKira acts on.

A ping, a start, and a failure

The success URL is the one the job calls when it is done. Call /start when the run begins and /fail when it exits badly, with an optional exit code. A run that is still open past its time limit counts as failed. Only a hash of the token is stored.

Late is not the same as one missed second

You set how often the ping should arrive, from every minute up to every 24 hours, and a grace period after that. UPKira confirms the miss before it opens an incident, the same way it confirms a website failure.

What this does not do

UPKira does not run the job for you, and it does not read the job's logs. The /fail body is a short note, not a log stream. There is one probe region, so this is about the job calling in, not about checking the job from several countries.

Questions people ask first

Does a heartbeat work on the free plan?
Yes. Heartbeat is one of the check types included on Free, inside the monitor limit for that plan.
What if the job is still running when the next one starts?
Call /start at the beginning and the success URL at the end. A run that stays open longer than its limit is treated as failed.

SSL certificates · API checks · All features · Pricing · Getting started