What do you use for offsite backup? Since best practice recommends 3 copies on 2 different devices where one device is offsite.
I thought about renting a storage box from Hetzner to use as an offsite backup but I was curious what you are using. And also if there might be some cheaper alternatives to my proposed solution that are equally as easy to setup.
You must log in or register to comment.
Backblaze here.
Backblaze is going to be the cheapest for pay as you go s3 as far as I’ve seen
Rsync.net with borg
I have a raspberry pi with a cheap 5TB usb drive at my parents house that boots up once a week and pulls a backup. I use rsnapshot to create incremental updates that takes up every little space and is easy to manage. I have the drive accessible with smb should I ever need to pull a copy from there. It’s super slow but that doesn’t matter for an off-site backup and it is super cheap
Edit: I should maybe add for future readers that cheap does not mean cheap quality but cheap relative to the amount af TB you get per dollar. I use a WD shingled drive wich is quality drive but cheap and slow af. But it doesnt matter because the internet connection is the bottleneck anyway.
Literally the same thing.
Every once in a while I have to have the onsite staff turn it off and back on again. (Aka my mom)
Thanks mom
I have a Synology NAS at my parents place and I backup there every night with borg
Acronyms, initialisms, abbreviations, contractions, and other phrases which expand to something larger, that I’ve seen in this thread:
Fewer Letters More Letters NAS Network-Attached Storage Plex Brand of media server package SATA Serial AT Attachment interface for mass storage
3 acronyms in this thread; the most compressed thread commented on today has 9 acronyms.
[Thread #277 for this sub, first seen 12th Nov 2023, 15:10] [FAQ] [Full list] [Contact] [Source code]
I’ve also been using hetzner storage boxes. They are as cheap as it gets and my internet connection is the limiting factor anyway.
I’ve recently started using a Hetzner storage box for encrypted daily incremental backups and I’m very happy with it so far.
I just keep an encrypted hard drive at work and bring it home to sync once a month/sooner if there’s a bunch of vacation photos we just added to the server or something. External hard drives are cheap nowadays
This is also what I do, weekly. It’s one of the cheapest (cheap SATA drive and USB enclosure, pay once) and most reliable methods, and arguably one of the most secure (the offsite backup drive is also offline most of the time).
A simple script on my desktop sends a desktop notification reminding me to plug the USB drive, once it’s mounted backups get pulled from my servers to the external disk, then I get a notification to unplug the drive and store it away. There’s about 15 minutes every week where all backups are in the same place. To be extra safe, use 2 drives and rotate them each week.
If you have a Windows or Mac machine, Backblaze will give you UNLIMITED backup storage (not B2) for $7 per month. They won’t let you use Linux and they won’t let you back up network drives, because that’s easy to abuse.
So, I have an 8TB drive in my Windows Plex server and shared on the network, and I have every other machine in my house backing up to that network location. Because the drive is local to that one windows machine, Backblaze will back it all up, and any other drives I put on there. I use FileHistory to back up my Windows gaming machine and SyncThing to back up non-Windows machines.
It takes a little work to set up with SyncThing, but I’m pretty sure that’s one of the cheapest ways to back up a shitload of data. And Backblaze recently upgraded the unlimited plan to 1 year of history.
On the flip side the B2 storage is great, it’s now even an S3 clone
True; B2 is definitely worth the extra money that you would pay for it, and my hacky way of exploiting their consumer backup plan is arguably a sketchy solution, and a waste of my own time. But I’m a cheap-ass, so that’s how I roll. 😆
You can try for linux: https://hub.docker.com/r/tessypowder/backblaze-personal-wine
That’s a neat option, thanks.
I use backblaze B2. I use duplicity to create a local encrypted backup of daily and then monthly incremental backups that are stored on a separate hard drive as a local backup. I then sync that with backblaze every night. It’s worked like a treat. Gives me my primary data, a local backup on a separate drive and then an off-site backup. And actually my primary data and my local backups are both on ZFS raidz2 drives, so I can even have drives fail and be okay.
I used to use glacier, but the backblaze interface and uploading scripts were just so much easier to use, and the price was comparable if not maybe just slightly cheaper, I can’t remember exact. I think duplicity also has a front end, duplicati that some people use, but I’ve never used it.
I’m using restic and Wasabi.
Are there any backup systems using object storage yet? That seems cheaper than file shares.
I haven’t used it, but restic was mentioned in an earlier backup discussion. It appears to be able to use rclone, which can talk to object storage.
Yep, currently using restic with Backblaze B2 via rclone
I use S3 with Kopia and some rules that move things to cheaper tiers as they get older.
My NAS does a backup to Synology C2 every night.
Filen
Onedrive /google drive for immediate stuff. Other stuff (too big for cloud services) from local to Synology, or simply served from Synology. Cloudsync from OneDrive/Google drive to Synology. (Periodic verification that things are sync’d this is very important!). Snapshots on Synology for local ‘oops’ recovery. Synology hyperbackup to Wasabi for catastrophic recovery. (used to use Glacier for this but it was a bit unwieldy for the amount of money saved - I don’t have that much data)
I’m aware that the loopback from onedrive/Google drive to synology doubles network traffic in the background but, again, I don’t have that much data and a consistent approach makes things easier/safer in the long run. And with more than one computer sharing a cloud drive link, the redundancy/complexity is further diminished. (let the cloud drive experts deal solving race conditions and synchronization/concurrency fun).
This works because every computer I have can plug into the process. Everything ends up on Synology (direct or via onedrive/Google drive) and everything ends up off site at Wasabi.
I very rarely need to touch the Wasabi stuff (unless to test, or because of boneheaded mistakes I make (not often) while configuring things.
It’s a good model (for me), adapts well to almost every situation and let’s me control my data.
