Maybe have a look at borgbackup docs to use the same terminology, otherwise it might get confusing.
When you create a backup ARCHIVE in a borg REPOSITORY, the archive contains all input files you gave to borg.
Each archive has a name and usually the name is something like machinename-setname-date-time.
You can create such backups rather often (that is cheap, due to deduplication), but you do not want to keep tons of archives long term (some borg commands take O(archive count) time).
Thus, one runs borg prune to "thin out" archives and only keep some following some prune policy, like keeping 60 daily, 12 monthly, 10 yearly or so.
It is important to use --prefix if one creates multiple different archive sequences in the same repo, e.g. if you create pc-home-date-time as well as pc-system-date-time - then you need to run prune with --prefix=pc-home- and another prune with --prefix=pc-system- .
Usually, archives are not modified after their creation (the only way to do that is to carefully use borg recreate command).
When you create a backup ARCHIVE in a borg REPOSITORY, the archive contains all input files you gave to borg.
Each archive has a name and usually the name is something like machinename-setname-date-time.
You can create such backups rather often (that is cheap, due to deduplication), but you do not want to keep tons of archives long term (some borg commands take O(archive count) time).
Thus, one runs borg prune to "thin out" archives and only keep some following some prune policy, like keeping 60 daily, 12 monthly, 10 yearly or so.
It is important to use --prefix if one creates multiple different archive sequences in the same repo, e.g. if you create pc-home-date-time as well as pc-system-date-time - then you need to run prune with --prefix=pc-home- and another prune with --prefix=pc-system- .
Usually, archives are not modified after their creation (the only way to do that is to carefully use borg recreate command).