Pinging & status
The ping URL is your check's heartbeat. Here's how to use it.
Signal success
A plain request to the ping URL — GET, POST, or HEAD — records a successful run and resets the check's clock.
curl -fsS -m 10 https://api.numo.sh/p/your-check-tokenSignal failure
Add ?status=fail to mark the check down immediately — useful when your job knows it failed and you want an alert without waiting for the schedule to lapse.
backup.sh || curl -fsS -m 10 "https://api.numo.sh/p/your-check-token?status=fail"Signal a start
Add ?status=startat the beginning of a long job to record a heartbeat without changing the check's state. Pairing a start with a later success lets Numo measure run duration.
Timeouts & retries
Keep your ping resilient but non-blocking: a short timeout (-m 10) and a few retries (--retry 5) so a transient network blip doesn't drop the ping — but the ping never holds up your job.
Grace period
Each check has a grace period — how late a ping can be before Numo marks it down. Set it a little above your job's normal jitter so a slightly-slow run doesn't page anyone. A check moves through up → grace → down as time passes without a ping.