Open 66krieger opened 1 year ago
I also use your blueprint and notice that there are always outdated backups in Supervisor. These backups (Hourly) were created by AutoBackup, but the slug name is not included in the file /config/.storage/auto_backup.snapshots_expiry
.
I've been observing this for a few weeks now and have absolutely no explanation for it. It is not fundamentally due to the backup type, because other hourly backups of Auto-Backup will be removed.
I see no pattern, no logic and no dependence on anything (e.g. manual deletions, backup errors).
Same. It seems to happen more when another service creates a backup (e.g. when you're offered to make a backup before upgrading part of HASS) and then this AutoBackup seems to lose track of backups prior to that. I often have to go and manually delete them.
Same on my install. Old backups are not being removed as per the "keep_days" flag.
same issue. these days almost no backups are deleted anymore. every couple of weeks i need to manually cleanup when the disk gives me a low free space warning. would be great to get this working again.
Same error. I create the backups with Node-Red, they succeed. I run the purge before every new save (they are set to 2 days). Unfortunately, it doesn't delete anything, it doesn't monitor anything, it just makes new ones.
Same Problem here: I use the Automation examples from: https://jcwillox.github.io/hass-auto-backup/advanced-examples/
(Full backup every day except Mondays, Full backup every Monday) to an external NAS, but purge seems not to work (already some backups deleted with find ... -mtime ...):
I can't say I've every experienced this issue, and I've been taking 3-hourly, daily, monthly backups for years now and it's never failed to track any. So the functionality definitely works, but the question is why not for you. There are a couple options other than it being a bug.
The deletion is handled by home assistant, so if you enable info logs and see purged being called then it's most likely out of our hands and into home assistants (or rather the supervisor), it could fail to find the backup with the slug, i.e. file was moved, or have some other file IO error. (hopefully you'd see some logs)
If you're not seeing the backup in the auto_backup.snapshots_expiry
then it's not being monitored, either it didn't receive the keep_days
(intentionally or misconfigured) or auto backup failed to get the id of the newly created backup, if your backups are taking over 20 minutes auto backup will timeout and not track the backup (it will still get created), all you can do is make smaller backups or increase the timeout in AB settings, but the longer the backup takes the more likely the request could timeout, meaning AB doesn't get the id. Additionally, if you restart home assistant during a backup, AB won't know when the backup is completed.
The problem
Hi, The backups made are not deleted, despite the setting. Where else can I look for the error? Sören
What version of Auto Backup has the issue?
1.4.0
What version of Home Assistant are you running?
Home Assistant 2023.7.3
What type of installation are you running?
Home Assistant Supervised
If you're running HA OS/Supervised, what version of the Supervisor are you running?
No response
Example YAML snippet
Anything in the logs that might be useful for us?
No response
Additional information
No response