How do I monitor a scheduled task (cron)?

Steps

1. Name the job, for example “nightly backup”. The name appears in alerts and reports. Say how often the job runs, and set a grace time: how long after the expected time the app waits before it calls the ping missed.
2. Copy the ping URL when it is shown. It is shown once and only a hash of it is kept, and anyone who has it can send pings for the job. If you lose it, make a new ping URL on the check.
3. Call the URL at the end of the job, for example at the end of the cron line:

curl -fsS --retry 3 <your ping URL>

1. Optionally call the same URL with /start when the job starts, so the run time is measured, and with /fail when it fails. You can also set a longest allowed run: with start and finish pings, a longer run counts as a problem.

Good to know

• Scheduled-task checks are counted per organisation, and the plan sets how many you can have.
• A scheduled task is confirmed by the ping receiver, not by two regions. A status page shows such an incident as a task that did not report in on time.
• It tells you the job did not report. It cannot tell you why; that is in the job’s own logs.