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.
VPS backup & restore
I have a VPS with a provider that includes backup/restore functionality with the service.
For the first time, I restored an image and the result is a non-functioning server.
If a backup is taken of a fully functioning system, shouldn't the system be fully functional when that backup is restored?
I feel like I'm getting the run around because the provider states that the backup and restore were successful because the server boots and can be logged into. They said it's not their problem that the server is not functioning as it was when the backup was created.
Is this acceptable? What's the point of offering backups if you can't restore your server to a working state?
Comments
It depends on what exactly is broken and it depends on what type of backup it is.
Backups (should) ensure that data is retrievable in case of a disaster.
Backups do not necessarily contain system state and cannot take external factors into account (such as a change of IP), so it's not entirely unreasonable for there to be some work required (by the user) to bring the server back to an operational state.
What exactly happened?
What is and isn't working?
Was there data lost?
I run Netbird on the VPS. I made several changes. Rather than undoing those changes, I restored a backup thinking it would save time. I do this all the time with the VM's on my LAN with Veeam B&R. It never lets me down.
Getting a 404 when accessing the Netbird admin and none of the clients can connect.
Not sure about data loss. Everything seems to be there.
So does OVHcloud or any other VPS provider include full snapshot copy of the entire system state so one can simply restore a backup and expect everything to be restored and functioning just as it was at the time of backup? If not, then what are the contributing factors as to why this cannot be done with VPS providers?
I use Veeam B&R on my network and when I restore a VM, it restores the system to the working state of when the backup was taken. It always works. This can't be done with VPS providers?
Most likely the backup is fine, and docker or some service just didn't start after the restore. Check docker ps and the service status.
If all the data is there, you probably just need to restart some services. IP change, time change, cache expiry, et cetera. I can think of lots of little reasons why it doesn't just work like a time machine.
I don't think I've ever used a VPS backup the way you describe. I can see how intuitively it might seem like it should just work. But a backup doesn't necessarily work like that. I can see how restoring from a backup would still require user intervention to get it back to an operational state. Unless you're paying for a managed server, I don't think it's an unreasonable thing to expect the user to know how to do.
Thanks for all of the responses.
This is actually the first time I've tried to restore a VPS backup. All of my backup/restore experiences have been with Veeam and server VM's on my LAN.
Below is what I get on the VPS:
A few possible things come to mind:
If the machine boots and filesystem etc seem to be OK, technically you have a restored backup. The problem is either something in configuration or the backup wasn't a state/snapshot backup - something Veeam in most cases is
Yes. Several times and it seems to be OK except for Netbird.
No.
SQLite. Based on the back and forth emails from the VPS provider, I'm assuming they use snapshot backups.
I have deleted the containers and re-pulled the images. Get same results.
Well, first you have the dashboard container restarting, start by checking its docker logs to see why, then with this up you maybe can see some netbird-specific diagnostic on that dashboard. I would check all the other containers logs for errors too.
This is probably some database hickup - often when data is being written when a snapshot is taken, the data is corrupted and would be restored by a backup or checked/fixed by some db command (like mysqlcheck for mysql). The system snapshot can only guarantee that it restores exactly was on the disk at that point in time, but the running programs of that time are now gone.
Nothing returned when I attempted to access the logs. Maybe my command wasn't correct. I spent a lot of time before giving up. I reinstalled the OS and Netbird.
This is what I get now:
first concern is how's ur data.
Don't rely on the provider's backup. Imagine they terminate ur service, data is also gone.
Thanks everyone for the comments.
I was finally able to get Netbird and Veeam to play nicely together. Using the overlay VPN, I can now quickly and securely backup my remote VPS's with application aware processing and file indexing. I did a couple of full volume restores and had the VPS's up in no time without having to adjust any VPS settings and without worrying about database glitches.