Skip to main content

Monitoring

A home instance does not need a monitoring stack. It needs four things you can do in a terminal: check it is alive, read its logs when it is not, keep working backups, and get told when something breaks while you are not looking.

  • Health and status: the health endpoint, the container healthcheck, and what a stuck deploy looks like.
  • Logs: reading docker compose logs per service, and what to look for.
  • Backups and restore tests: the daily dump, what to keep, and proving a backup actually restores.
  • Alerting: wiring a webhook so a problem reaches you instead of waiting to be noticed.

Cobblr makes no unsolicited phone-home: nothing here reports to an outside service unless you set it up. Monitoring a self-hosted instance is watching your own box with your own tools.