Run it from cron.
A schedule is five cron fields saved on the site. BackupBinder stores the fields. Your system's cron is what actually starts the run.
Turn the schedule on
The change needs the site's id and its exact root. The default expression is 15 3 * * *, which is 03:15 every day. --no-schedule during setup leaves the schedule off.
./bin/backupbinder set_schedule --id SITE --root /path/to/site --cron "15 3 * * *" --enabled
Saving the fields does not start a backup by itself. Something on this machine has to call the program.
The cron line
Point system cron at the binary once a minute. run_scheduled looks at each site and backs up the ones that are due. It calls the same path as create_backup. Nothing in that path stops to ask a question.
* * * * * /path/to/backupbinder --home /path/to/var run_scheduled
doctor prints this line with your binary path and data directory filled in. It still does not install the line. inspect_site shows the cron fields saved on one site.
How many snapshots to keep
Each successful run is a restic snapshot. Unchanged file data is stored once. A new site keeps the newest 3 snapshots, 24 hourly, 14 daily, 8 weekly, 12 monthly, and 2 yearly. A snapshot stays if any of those rules keeps it.
After every backup, snapshots outside the rules are forgotten and the freed space is pruned, in the local repository and on each destination that just received a copy. apply_retention does the same pass for every enabled destination.
./bin/backupbinder set_retention --id SITE --root /path/to/site --keep-daily 14 --keep-weekly 8
./bin/backupbinder apply_retention --id SITE --root /path/to/site
0 skips that one rule. Set every count to 0 and every snapshot is kept. Leave a flag off and that count stays as it is. The flags are --keep-last, --keep-hourly, --keep-daily, --keep-weekly, --keep-monthly, and --keep-yearly.
Saving the rules applies them immediately when a repository already exists. If there is no repository yet, they apply on the next backup. The same form is on the site page.
Slow the copy down
A site can cap restic repository traffic. The numbers are KiB per second, matching restic --limit-upload and --limit-download. 0 or a blank field means that direction stays unlimited.
./bin/backupbinder set_limits --id SITE --root /path/to/site --limit-upload 1024 --limit-download 1024
The cap applies to backup, copies to other destinations, checks, and restores, including a scheduled run. The same fields are on the site page.