New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Share your backup strategy.
ReliableSiteHosting
Member, Patron Provider
in General
What's your backup strategy?
Comments
don't
i just trust my provider to handle that
Always backup after losing the data
Regular DB dumps & file backups to Backblaze B2 (everything) and Cloudflare R2 (only latest)
For my most important server I have a bash script that sets it all up, it’s way too big but I ain’t fixing that
For the things I’ve set up more recently i.e last few years, everything is Ansible so restoring is very easy
I buy from reputable providers with automated backups.
If none, I don't buy.
Btrfs snapshots for now.
3-2-1
Three copies of my data on two different types of media of which one is off-site.
Different types of media can also be one copy on a storage server and another one with a cloud provider.
Some advice:
My provider includes free snapshot backups, but I don't rely on those alone. I also keep copies on a separate storage VPS using SFTP for extra protection.
Real men don't do backup, trust their providers and their instinct
3-2-1 - a mirrored system (zfs > btrfs) + external backup + remote
Borg backup is very good - encrypted backups that work like an image mounted over FUSE - works fine with a DirectAdmin Storagebox
Borg is awesome.
Right now the most complete backup solution I have is of my minecraft smp server with 0 daily players. It's more of a system admin habit than not. Here is how:
As for images on my phone, it is immich on my VPS exposed publicly through pangolin(public yet almost majority of things are blocked).
For manual homelab stuff like experimenting new scripts and shit, actual data is stored on storage box and config locally which i backup every once in a few months manually.
For all of my backup solutions, a restoration has been tested atleast once for each component. Especially considering how often I break shit up.
I have simple rsync for these backup strat, syncing every minutes
4 dedi with raidz1 array on EU (spread out production)
1 dedi with big 2x raidz2 array on NA (backup location for 4 dedi as well acting as failover)
offline cold backup (multiple 8tb drives) on my house, automatically synced every week
Doing my best to:
P.S.
I don't have proper email backups yet.
Would love to hear what you guys do to keep your inboxes easily restorable.
rclone to storage vps.
Very easy: Just don't loose data. Problem solved.
What's backup?
7x daily backups to remote server
Restic
Amazon S3 , Heztner storage box
Most of it uses restic, I find the restic rest-server api just so convenient for remote backups plus it really is an awesome piece of software.
3-2-1 strategy most of the time, have a couple of storage servers with different providers that I rotate as final target.
321 the hard way. Paid 800 to recover 2 hdd raid 1 becoz both degraded n i was sleeping. Now 1 copy daily sync home n 1 copy weekly sync remote s3
1 onsite and 2 offsite Proxmox Backup Servers. And a lifetime rsync package for Zerobyte to backup Immich photos.
Rsync to Hostbrr, rclone to backblaze and storj
We're not allowed to back up in Vermont so K turns only. Good luck with the leaf peepers.
(B2 for all but the stuff that I can't afford to lose period, which doesn't leave my house. If jurisdictional arbitraging is needed mostly from the travesty that gives Article II administrative agency basically unchecked copyright interpretation and enforcement power that should not even be on the table, they're in countries that don't exist or the FBI can't lease an office in but is not under sanction.)
Depends on the workload.
For my Proxmox infra, I use PBS for VM and container backups, with raid redundancy.
For hosting panels like DA, cPanel I rsync usually to a remote location, storage box on hetzner or even my own controlled server, if thats raid backed the better.
For any standalone box I have my own cron jobs which just have served me well throughout, just dumping sql and archiving data and rsync off.
The only thing I always tell myself is RAID although nice, is not a backup strategy, its nice when your disk goes but it wont protect against deletion, corrupt data or even rootkit on your server or a bad update, so RAID is nice for disk redundnacy while backup is recovery.
I like belt and suspenders, sometimes I can go a bit anal on it, but it has served me well in the past being anal and having even emergency backup, maybe 1 backup a month for the past 12 months. Yes, you can often feel like "wow thats nuts" but for me it has worked well knowing that I can find a January backup before the issue occurred that I just located in May
My two cents, backup your backups ;D
My LET low end backup solution;
snapraid + qbackup + cmd/bash backup scripts + cron/windows task scheduler.
https://www.snapraid.it/
https://www.qualeed.com/en/qbackup/
Pbs backups for most things
Then for files i use 2-5 locstions of duplicati
Some files get cold storage once every few years
For LET services? No backups because each service is nearly identical. If it dies, I can reinstall and configure it identically to the way I had it before. For the few services that actually hold unique data, I just do a periodic rsync backup over ssh.
For my own important shit that I keep at home, I follow 3-2-1.
simply using restic backing up to amazon s3