On Tuesday, 4 May 2021 11:51:34 AEST Keith Bainbridge via luv-talk wrote: > > Traditionally the answers are: > > 1. read-only snapshots (or git)
For small systems I recommend BTRFS for snapshots and ZFS for large servers that have many fixed disks. > > 2. RAID RAID with checksums as the vast majority of RAID issues I've seen have been drives returning corrupt data and claiming it to be good. That means BTRFS or ZFS on Linux and ReFS on Windows. > > 3. offsite tape/usb-hdd rotation Cloud is an offsite backup option too. > situation. And the way I play with stuff, disaster is a wrong > keystroke away. (What happens if you > dd if=//.iso of=/dev/sda ?) You had better have a good backup of /dev/sda in that case. If you run ZFS or BTRFS in a RAID-1 array on /dev/sda and other drives then the above command won't even interrupt your work. Part of my BTRFS and ZFS training is doing exactly that and showing that no data is lost. A large part of the training is deliberately corrupting disks to see how things are recovered. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/ _______________________________________________ luv-main mailing list -- [email protected] To unsubscribe send an email to [email protected]
