--- title: "Cron job monitoring with heartbeat pings · Fettle" description: "Cron job monitoring with heartbeats: each job pings a URL when it finishes, and Fettle alerts you when a run is late or reports a failure." url: "https://fettle.sh/heartbeat-monitoring" --- / heartbeat monitoring # Cron job monitoring that alerts when a run is late or fails A heartbeat is a URL your cron job, queue worker or backup script calls when it finishes. Tell Fettle how often the job runs and how long to wait. When the ping does not come, or the job reports a failure, you get an alert before anyone notices the missing data. Heartbeats are on every plan, free included. The [cron job monitoring guide](https://fettle.sh/use-cases/cron-job-monitoring) walks through the setup step by step. [Start free](https://fettle.sh/register) Free plan: 20 monitors every 3 minutes from 1 region. No card. ## One line at the end of the job Create a heartbeat with the job’s period, copy its ping URL, and chain a request to it after the command. A plain GET or POST works, with no API key: Ping after a successful run (crontab) ``` 0 3 * * * /usr/local/bin/backup.sh && curl -fsS -m 10 --retry 3 https://api.fettle.sh/v1/heartbeat/ > /dev/null ``` The ping goes out only when the job succeeds, so a crash, a hung run or a server that never started the job all end the same way: a missed deadline and an alert. The [heartbeat reference](https://fettle.sh/docs/heartbeats) covers period and grace, the /start and /fail pings, and the create API. ## What a heartbeat catches - Late runs When a ping does not arrive within the period plus the grace you set, the heartbeat goes down and you get an alert. Fettle checks every 30 seconds. - Failures the job reports A job that knows it failed calls its ping URL with /fail and the alert goes out at once, without waiting for the deadline. - How long each run took Call /start when the job begins and the success ping at the end, and Fettle records the run's duration. - No alarm before the first run A new heartbeat waits for its first ping before it can be late, so creating one ahead of the deploy is safe. - Recovery alerts The next successful ping closes the incident and tells you the job is back. - From a script or an agent Create, pause and resume heartbeats through the [API](https://fettle.sh/api), or ask Claude or Codex to do it over the [MCP server](https://fettle.sh/mcp). ## Where the alert goes Heartbeat alerts use the same channels as your uptime monitors: email and Telegram on every plan, and Slack, Discord or a signed webhook from Plus. Maintenance windows silence monitors, not heartbeats: before turning a job off, pause its heartbeat. ## Plans Heartbeats per organization: 3 on Free, 20 on Plus, 50 on Pro and 150 on Max. They count separately from monitors. Every plan and limit is on the [pricing page](https://fettle.sh/pricing). ## Read next - [How to monitor a cron job with a heartbeat A cron job that stops running fails no request and throws no error. Here is how a heartbeat monitor catches the silence and alerts you before users notice.](https://fettle.sh/use-cases/cron-job-monitoring) - [Heartbeats: ping URLs, period and grace How Fettle heartbeats work: create one, call its ping URL from a cron job, report start and failure, and set the period and grace that decide lateness.](https://fettle.sh/docs/heartbeats) - [SSL certificate monitoring and expiry alerts SSL certificate monitoring: an alert days before a certificate expires, an incident when the chain or hostname breaks, and expiry dates in one list.](https://fettle.sh/ssl-monitoring) - [Status page that updates from your monitors A hosted status page fed by your uptime monitors: live component status, daily uptime, incidents that open on their own, and email updates for subscribers.](https://fettle.sh/status-page) [Start free](https://fettle.sh/register) Free plan: 20 monitors every 3 minutes from 1 region. No card.