Duplicati: flag warnings run after run, and forecast how much room backups take
A job whose last three runs in a row all finished with warnings is flagged and raises an alert (on by default, switchable off per job). The size of the destination, and its quota where the storage says, are read from the reports and kept with each run; the Stored column shows the growth and, where the room is known, when it will be full, with an alert inside the disk-forecast window. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
This commit is contained in:
1 parent
333146e8e3
commit
4f077d4f4a
15 files changed
+10115
-14
No files matched your search
@@ -251,6 +251,8 @@ Duplicati tells this app how each backup went ([INTEGRATIONS.md](../INTEGRATIONS
|
||||
The alert says which job, on which machine, why (the first error Duplicati reported) and when the last good backup was. Only the *Backup* operation counts: a compact, a repair or a verification does not say whether the data is safe. A machine that is heard but has no jobs says why on its card (the last report was another operation, or could not be read, with counts since it was added) — see [INTEGRATIONS.md](../INTEGRATIONS.md#duplicati).
|
||||
A job can be **muted** (no alert; it is still listed),
|
||||
**Alert on warnings** (an operator's button on each job, off until switched on — some jobs always have a few, and a backup with warnings is still a backup) raises a warning-level alert while the job's last run had warnings, naming the job, how many there were and the first thing they said; the next clean run ends it. A job that is muted raises nothing, this included. Maintenance on the server silences it like the others.
|
||||
Two slower things are watched too. **Warnings in a row:** when the last **three runs in a row** all finished with warnings, the job says so (*Warnings 4 runs in a row*) and a warning-level alert is raised (`duplicati-streak:<job>`) — on by default, off per job with *Stop streak alerts*, and not raised for a job that already alerts on every warning run or that has failed or is overdue. **Room:** Duplicati says how big the backups are where they are kept and, for some storage, how much room there is; the *Stored* column shows the size and its growth over the last 180 days (once there are five readings over a week), and where the room is known an alert (`duplicati-growth:<job>`) is raised when the storage will be full within the **disk forecast window** (Settings → Alerts, 14 days; 0 = off) and the growth is steady; critical at three days or less. Storage that does not say how much room it has is never forecast to fill.
|
||||
|
||||
A job can also be muted, and one that no longer exists can be **forgotten** (with its history; if Duplicati reports it again it is learned again). Each machine can be tied to a server, so **maintenance on that server silences its backups' alerts**.
|
||||
|
||||
What is kept: the last result of each job and what its warnings said (up to the first twenty, each cut at 500 characters — Duplicati itself may send fewer than it counted, and the page says so), and each report, with what its warnings said, for 400 days or 400 reports. `/metrics` has `homelab_duplicati_backup_ok`, `homelab_duplicati_last_success_age_seconds` and `homelab_duplicati_source_size_bytes` for each job. A job is only learned from its first report, so a job that has
|
||||
|
||||
Reference in new issue
Block a user