Linux Install and Backup Questions

I’m the only person in my family that is into computers. I’m thankful that I’m not anyone’s IT guy, though. :laughing:

If my old AM3+ motherboard hadn’t died, I actually would love to make that a project.

1 Like

I wont go out to spend on unnecessary computer parts for now if all works as it is. It’s not currently practical with the current chip shortage.

If you have close friends, however, maybe try to pester them if they have unused hardware and ask if they can permalend it to you (or buy them cheap). :rofl:

1 Like

No friends that are into computers either. I’m a black sheep. :rofl: There is a local computer shop, though, but this would definitely be a project for the future if I can make the room for another tower.

I would REALLY prefer going rackmount and going vertical, but money constraints. :slightly_frowning_face:

I’ve just gone into rackmount and while i dont necessarily regret it, the only real benefit of rackmount is saving space. If space is not an issue, you should focus more on the stuff running on the rackmount rather than the rackmount setup itself (if that makes sense).

1 Like

You mean going open rack instead of enclosure rack?

I have a ton of parts bookmarked for a dream multi-server rack set up, but alas, 'tis but a dream. lol

I mean focus on the software and actually running the server and its hosted apps rather than focusing on the hardware components. Even a cheap homelab with Raspberry Pi 4 should be sufficient (unfortunately with the chip shortage, even RPi4s are insanely priced).

1 Like

I would actually love that. I’m interested in this thread because I’m in a similar position to op. My system is a hodgepodge of drives I’ve had going back a decade now. I have a 256GB m.2 boot drive; a 1TB and 512GB SATA SSD for game installs; a 2TB mechanical hard drive for general storage, photos, documents, etc; a 2TB m.2 drive for video rendering, a 12TB mechanical drive for films…

At the moment I do use them all for different things because they’re different speeds and sizes and therefore better optimised for different purposes. I mounted various subfolders of my /home folder across all of them using fstab so it’s presented in a way that makes sense, but my dream would be to have a tiered system that automatically and dynamically moves and allocates files to the drives that most sensibly suit the purpose and frequency of access. I’ve seen that ZFS and BTRFS can do this but honestly I am far too much of a noob at the moment to make head or tail of what I’ve read.


There are differences in the way distributions implement swap, so I am sorry to say “it depends”.

If you are using Ubuntu or something related to it have a look at the FAQ on swap in the Ubuntu docs . A search for “ubuntu swapfaq” should bring it up. It looks sane to me

If you plan to use Fedora or something similar, try searching for “red hat Recommended System Swap Space” and you should find a similar page for Red Hat and related distros.

YMMV applies to the following. I have a swap partition that is the same size as the RAM because I don’t use hibernation. The swap partition is barely used (< 1%) on this system.

1 Like

I’m heavily leaning toward an Arch related distro, so I’ll check their Wiki. Thanks for that info.

A guide would be fantastic.

I’m taking the suggestions from this thread and running a multi-disk VM to try everything out. I’m a noob with RAID. Though I fully understand all of the concepts, I’ve never actually put it into practice. I know I don’t want RAID 0, though, solely for the drawback of no redundancy. I could care less about the extra speed.

I’m the opposite I don’t care about the redundancy xD i think it lulls people into a false sense of security thinking it’s a backup, and before they know it their RAID system has replicated errors across their entire array. I’d much rather have a Raid 0 and back up anything I don’t want to lose myself.

1 Like

That’s fair. I don’t understand how a lot of people don’t understand that it’s not a backup. Whenever I was researching RAID years ago that was VERY heavily stated over and over, so I’m thinking those people just skim rather than actually research.

I think “RAID is not a backup” is a phrase people hear and regurgitate a lot, but it’s pretty rare for anyone to ever go into why it’s not a backup and how data loss still occurs in spite (or sometimes because) of RAID. That’s the part that a lot of laypeople find counter-intuitive. It requires some knowledge of how the technology actually works, while the issue with RAID0 just requires a very basic understanding of probability.

1 Like

Very excellent point.

Oh it literally just occurred to me that, while this isn’t technically RAID it is redundant mirroring, I accidently pressed the wrong button on a steam controller and deleted a 100 hour save file last year. Which steam cloud immediately replicated. Steam cloud isn’t a backup either lmfao

1 Like

It’s why you also hear the 3 - 2 - 1 backup method mention a lot.

3 copies of your data on at least 2 different media and 1 copy off site.


raid-1 is not a backup - when you delete a file by accident it just means it’s deleted more reliably :slight_smile:

On the other hand, snapshots can be backups, even if just using the same device.


I find Steam’s directory system completely unnavigable. It can be really difficult to find the local files you want to back up.

Oh definitely not. They’re a handy tool to have, but it’s no more a backup than storing a copy of all of your files on a directory next to the originals on the same drive. If that disk dies you lose it regardless.


Thing is with steam … I don’t consider backing up steam’s installed directory as 99.99% of the files can just be redownloaded onto a new system / drive. The saved game location is what I do back up.

As for what steam file for a game you want to back up just create a new steam library and move the game you want to the new folder thru steam and there are your files.

1 Like

The save game location isn’t always obvious.

1 Like