Build the snapshot with the below file

btrfs subvolume snapshot -r /path-to-sv/subvol /path-to-sv/subvol.ro

# send the subvolume to file, compress with parallel ZSTD & monitor progress
btrfs send /path-to-sv/subvol.snap.ro | \
    pv -c | pzstd -16 |  pv -c | \
    dd of=/path-to-external-backup/subvol.zstd.back

# delete read-only snapshot
btrfs subvolume delete /path-to-sv/subvol.ro

To restore subvolume from backup we run the process in reverse:

# read backup file and decompress the stream, redirect to temporary read-only snapshot
dd if=/path-to-external-backup/subvol.zstd.back | \
    pv -c | pzstd -d | pv -c | \
    btrfs receive /path-to-sv/


# make a RW subvolume 
btrfs subvolume snapshot subvol.ro subvol

# delete temporary snapshot
btrfs subvolume delete /path-to-sv/subvol.ro    

From here

https://superuser.com/questions/1396241/btrfs-imaging-a-volume-to-an-external-file

You might want to make this into a systemd timer to run at boot or before shutting down

https://linuxconfig.org/how-to-schedule-tasks-with-systemd-timers-in-linux

I’m lucky to have only had one system nuked by a faulty power supply that shut down during a kernel update.

I usually just reinstalled back then. But I didn’t get into it till the late nineties. Back when Ian was still on the list serves.

Unless you mean nuking the OS or borking the bootloader. Then yeah, countless.

Fedora based actually