After my QNAP 451+ died, what's the best backup strategy for a new Ugreen dxp4800 pro (Docker, VMs, Data)?
After my QNAP TS-451+ died (3x20TB RAID5), I had to decide on a replacement — build my own or buy a finished product. Given the current market situation, I went with the Ugreen DXP4800 Pro.
So far I really like it, but there are a few things bugging me that I want to sort out before committing fully.
Docker backup:
The Ugreen runs a Portainer agent, with Ugreen providing the /volume1/@docker directory for it. Volumes managed by Portainer can NOT be backed up with Sync and Backup.
You can work around this by using "Bind to Host" in Portainer, which I'd rather avoid. Or there's this: https://github.com/Railsimulatornet/UGREEN-NAS-Docker-Backup-Restore
VM backup:
Same problem as with Docker — VM disks are stored under /volume2/@kvm, and you have to manually back them up to get the .ova + .qcow2 files. Not ideal if you're running a Proxmox Backup Server.
Data backup:
This is the dealbreaker for me right now. I know BTRFS has a versioning feature, which is great, but in Sync and Backup you can only choose between:
- Incremental backup — keeps everything, nothing gets deleted until the drive is full
- Image backup — a 1:1 copy
Neither is really a "backup" in the proper sense — it's more like a sync with no real data retention.
On top of that, the QNAP's RAID5 setup (mdadm for RAID, LVM2 for the pool, ext4 + QNAP's proprietary thin provisioning) made it extremely hard to get my data back — 5 days of downtime plus 3 days to restore. I'm just glad I had a working backup and an offshore sync of my most important data.
So now I want to rethink and optimize my backup strategy.
A NAS really only has 3 core jobs: automated backup of Docker containers, VMs, and data.
Has anyone else run into similar concerns, and how are you handling it? I'm currently considering switching to TrueNAS Scale, since my next NAS will probably be a self-built one anyway, should the hardware market recover.