Healthchecks

Healthchecks watches the jobs that fail quietly. Each cron job or scheduled task pings the service over HTTP or by email when it runs, and when a ping does not arrive on time, the alerts go out. A backup that stopped working three weeks ago stops being a surprise.

A check carries a period, the expected time between pings, and a grace time saying how long to wait before treating a late job as a failed one. Schedules can also be written as cron expressions, parsed and evaluated with the cronsim library.

Around that core sits the machinery a team needs to use it daily.

• Live dashboard: a checks screen showing the status of every job and updating as pings arrive.
• Notifications: more than 25 integrations for delivering alerts, plus monthly email reports.
• Team management: projects, team members and read-only access for people who should not change anything.
• Status badges: public but hard to guess addresses you can drop into a README or a status page.
• Account security: WebAuthn two-factor support on the accounts that hold all of this.

It is a Django application on Python 3.12 or newer, storing data in PostgreSQL, MySQL or MariaDB, with SQLite as the default for a development checkout. A Dockerfile and pre-built images are published, the license is BSD 3-clause, and a hosted service exists for anyone who would rather not run a server.