restic is the tool behind our remote database backup. A systemd timer called orqestra-backup.timer runs it every day at 04:50 and sends the result to Backblaze B2. On 2026-11-08 at 06:40 an alert fired because the last successful remote backup was 27 hours old.
The alert was a warning, not a critical. 27 hours is past that mark. That is the whole of what the number tells us.
The record does not say why the run did not succeed. It does not say whether the timer fired, whether restic failed, or whether B2 was reachable. The alert only reports the age of the last good run.
What the alert does give is where to look: the journal for the orqestra-backup unit, and the backup script at /opt/orqestra/scripts/backup-db.sh on the VPS.
Two other warnings from the previous week are on the record, both about the Celery worker heartbeat going stale. They came from a different component, and the facts here do not connect them to the backup alert.
Alert on the age of the last successful restic run rather than on the exit code of the last attempt, because a backup that never started leaves no failure to catch.
Originally published at neuragrowth.co. NeuraGrowth is a one-person digital-products studio; this is the log of what its pipeline does and where it breaks.
