There's something about prune I must not understand

Getting Help
1

This is from a Windows environment running Restic 0.18.
I run a weekly prune command on my repository to delete old snapshots and hopefully keep my repository as small as possible. I use this command to do that:
restic.exe forget --keep-within-daily 7d --keep-within-weekly 1m --keep-within-monthly 6m --prune >> $outlog 2>&1

Here’s a snippet from my most recent forget/prune command run on 7/20:
keep 5 snapshots:
ID Time Host Tags Reasons Paths Size

ce222bfe 2024-09-10 14:21:14 HessNet-HS oldest daily within 7d D:\ServerFolders\Documents 11.609 GiB
oldest weekly within 1m
oldest monthly within 6m
2f752e9f 2024-09-10 14:32:08 HessNet-HS daily within 7d D:\ServerFolders\Documents 11.609 GiB
e0b8fa3c 2024-09-11 02:00:03 HessNet-HS daily within 7d D:\ServerFolders\Documents 11.609 GiB
bc8b9266 2024-09-12 02:00:04 HessNet-HS daily within 7d D:\ServerFolders\Documents 11.609 GiB
318bd7a7 2024-09-13 02:00:02 HessNet-HS daily within 7d D:\ServerFolders\Documents 11.609 GiB
weekly within 1m
monthly within 6m

Why would forget be keeping snapshots from last September? It seems like they should be deleted because they’re outside my 6 month window. Am I missing something?

2

" If there are not enough snapshots to keep one for each duration related --keep-{within-,}* option, the oldest snapshot is kept additionally and marked as oldest in the output (e.g. oldest hourly snapshot )."

Potentially that?

3

Thanks for the explanation. That’s logical.

4

You may check snapshots’ paths. By default restic forget groups snapshots by host+paths. If you change backup directory list, restic forget will process these snapshots as separated groups and apply –keep-within* to every group.

Also you may run restic forget … –group-by=host –dry-run and check the difference.