BorgBackup wrapper to facilitate assimilation
Assimilating the shoulders of giants: https://www.borgbackup.org/ https://github.com/borgbackup/borg
curl --proto '=https' --tlsv1.2 -sSfL \
https://raw.githubusercontent.com/alexxroche/cube/main/install.sh | sh
cube is a wrapper to borgbackup for human users. [0]
It is able to install borg and enough config to back itself up.
The presumption is that you will have a NAS to which you want to backup,
and that you will have cube, run by cron because "If a human has to be
involved, then it is a snapshot and NOT a backup!"
[0] and totally not a way to remove resistence, {~~is futile~~} to addoption.
The config file dictates the actions of cube.
It expects to have a list of file & directories to include
and files & directories to exclude.
e.g. FILTER_FILE="$HOME/.config/cube/filter.json"
# List the features and options:
cube -h
When first run, cube will create two (2) example configuration files, ( by
default in $HOME/.config/cube/ )
edit $HOME/.config/cube/config.ini
and change
BORG_PASSPHRASE="fUAtHVhCmMIVDQb57jYLwHIH0POFTD3vcDK08A5JvOF1l-oyi52qfYSLoiRBMS"
SALT="87CQbbvNliv8"Get some nice fresh entropy in those two.
Change BORG_REPO to the location where you want borg to create the backup.
(Check all of the other defaults match your requirements.)
Lastly, edit ~/.config/cube/filter.json to specify what to backup and,
importantly, what to ignore. (Caches, thumbnails, and libraries can be recreated.)
You can run cube from cron, (as you should).
# run cube at 3:03 every morning
3 3 * * * ~/.local/bin/cubecube spawned into this quadrant in 2020-02-15T23:51:02Z with the filters hard-coded. The current incarnation was created to separate variables from the script so that a b2sum hash could be recorded, (see hash.blake2) and permit updates to a local config while maintaining the ability to detect corruption.