You are not logged in.

#1 2018-07-19 23:03:35

jamtat
Member
Registered: 2008-03-13
Posts: 224

System back-up/restore using rsync and confusion over wiki entry

What is the difference between the two types of system back-ups described at https://wiki.archlinux.org/index.php/rs … tem_backup and https://wiki.archlinux.org/index.php/rs … em_cloning respectively? Don't they do, effectively, the same thing?

As far as I can tell, the only possible difference is in preserving some additional (and perhaps not absolutely necessary) metadata. Though the cloning entry touts as a virtue the fact that "source and destination file systems don't need to be of the same type," doesn't that same feature apply to the full system backup method as well? My rather rudimentary comprehension of what is being described does not give me any inidication as to why a restored backup using the first method would require that the file system be the same as that of the backed-up system. What might I be missing?

I'm asking because I'd like to take, for insurance purposes, a sort of snapshot of my Arch systems, from which I could later restore the system, should I meet with some sort of catastrophic hardware failure. Wouldn't either or both methods enable me to do that?

Offline

#2 2018-07-20 01:59:58

waitnsea
Member
From: France
Registered: 2013-02-10
Posts: 57

Re: System back-up/restore using rsync and confusion over wiki entry

jamtat wrote:

What is the difference between the two types of system back-ups described at https://wiki.archlinux.org/index.php/rs … tem_backup and https://wiki.archlinux.org/index.php/rs … em_cloning respectively? Don't they do, effectively, the same thing?

As far as I can tell, the only possible difference is in preserving some additional (and perhaps not absolutely necessary) metadata. Though the cloning entry touts as a virtue the fact that "source and destination file systems don't need to be of the same type," doesn't that same feature apply to the full system backup method as well? My rather rudimentary comprehension of what is being described does not give me any inidication as to why a restored backup using the first method would require that the file system be the same as that of the backed-up system. What might I be missing?

Hi,
This post gave a good example cloning old disk to SSD new one
Differences - I think - is that for cloning you may copy every existent metadata unuseful in backup
Note that for incremental backups you need an --exclude list (from file is the best) and no for cloning
Restoring a backup you absolutely must use --delete option to avoid  duplicates in pacman's local database

jamtat wrote:

I'm asking because I'd like to take, for insurance purposes, a sort of snapshot of my Arch systems, from which I could later restore the system, should I meet with some sort of catastrophic hardware failure. Wouldn't either or both methods enable me to do that?

Instead of "Snapshots" as btrfs can do I discovered Borg which makes the best solution for as well System and Data
  Maybe someone would translate my french-wiki contribution to official Wiki  with a better english I could do, even assisted with DeepL ?

Last edited by waitnsea (2018-07-20 02:05:36)

Offline

Board footer

Powered by FluxBB