Q: Node Backup

bferrell

Well-Known Member
Nov 16, 2018
99
2
48
54
I backup my guests regularly, but after my little glitch this week, I was wondering (and I've searched) - is there any way to backup the host/node configuration?
 
Obviously, Snarky McSnarkbutt? Of course I looked in the forums. Just because I didn't find it doesn't mean I didn't look. Thanks for the link. I was hoping there was something built into the system, not some github scripting hack, but that's good to know.
 
Obviously, Snarky McSnarkbutt? Of course I looked in the forums. Just because I didn't find it doesn't mean I didn't look.

Yes, you're not the first one that asks about that ... but still. Let's see this as learning opportunity and increase your google skills, because it's literally the first link.

https://www.google.com/search?q=self-backup site:forum.proxmox.com

upload_2019-7-14_16-9-27.png
Even with your title, it's the fourth link - at the time of your search it was the third.

upload_2019-7-14_16-12-12.png
 
S-u-p-e-r helpful, Dad... Oh, and I'll be late for dinner. Thanks!

Sorry, nothing personal and I don't want to be a schoolmaster, but claiming to have searched and be disproven so easily ... I had to jump the gun.
It's just a case of negative serendipity.

Thanks for the link. I was hoping there was something built into the system, not some github scripting hack, but that's good to know.

Besides all the ranting, I forgot to comment on the matter. As always - depends on what you want. You used the word 'node' to describe your host, so you're running in a clustered setup. In this setup the PVE configuration data is already replicated across nodes and will not be lost if one node fails. If you "just" want to have a backup of your node so that your time-to-recovery is minimal, you can just use any linux backup utility around, but in most cases, you end up with a homebrew backup solution like the github project. Linux has too many ways to do things and with regard to backup solution, this is very bad.
A homebrew solution depends naturally on your Linux knowledge, e.g. with ZFS as your backend storage, you can replicate via send/receive your whole system off-site and restore it by booting up a live Linux with ZFS support, restore your partition layout, create your pool and replicate your data from the off-site backup back to your host. We use that approach on our single-node off-site servers a lot.
If you're working in a larger company and/or love automation tools (ansible, chef, puppet, etc.) you can just provision your machine automatically back and restore the necessary files for cluster integration and are golden.
Others use etcd to store the host configuration off-site (via git push)-
A point-and-click backup AND restore solution does unfortunately not exist out-of-the-box in PVE (patches welcome!). The only operating system I know that has integrated support for restore on install is MacOS (and I mean by the OS vendor, not some third party).

Final words on statistics, we obviously do also backups of the OS filesystem if necessary and wanted, but we did not encounter one failure in this decade on over 200 machines. All servers have at least two disks for the OS, some also 3 in mirrored configuration.
 
Yea, thanks, that's really helpful actually. My search skills might suck, but that's no crime, and you can choose to not reply rather than lambaste folks for having weak search skills. I'm not a unix guy really, and this is for my home setup; I'm an engineer but I don't do this for a living. I realize the noob questions can get really annoying, and I do try (believe it or not) to find my own answers. I'm not quite ready to go the scripting route, but I do get nervous when I have issues like this week with corosync changes so I'll look into when I get a bit more experience. I do appreciate the pointer.
 
Hello,
I was thinking of creating thread but there is no need now :),the reply i read at @DerDanilo post reply there is backup node so how to do that,I'm searching but there is no mention in proxmox docs.
then there is way to do that ?
and if there is not:( what prefer/suggested tools to do that ? what the type of preferred data format to backup to it ftp,nfs...?
thanks
Sorry, nothing personal and I don't want to be a schoolmaster, but claiming to have searched and be disproven so easily ... I had to jump the gun.
It's just a case of negative serendipity.



Besides all the ranting, I forgot to comment on the matter. As always - depends on what you want. You used the word 'node' to describe your host, so you're running in a clustered setup. In this setup the PVE configuration data is already replicated across nodes and will not be lost if one node fails. If you "just" want to have a backup of your node so that your time-to-recovery is minimal, you can just use any linux backup utility around, but in most cases, you end up with a homebrew backup solution like the github project. Linux has too many ways to do things and with regard to backup solution, this is very bad.
A homebrew solution depends naturally on your Linux knowledge, e.g. with ZFS as your backend storage, you can replicate via send/receive your whole system off-site and restore it by booting up a live Linux with ZFS support, restore your partition layout, create your pool and replicate your data from the off-site backup back to your host. We use that approach on our single-node off-site servers a lot.
If you're working in a larger company and/or love automation tools (ansible, chef, puppet, etc.) you can just provision your machine automatically back and restore the necessary files for cluster integration and are golden.
Others use etcd to store the host configuration off-site (via git push)-
A point-and-click backup AND restore solution does unfortunately not exist out-of-the-box in PVE (patches welcome!). The only operating system I know that has integrated support for restore on install is MacOS (and I mean by the OS vendor, not some third party).

Final words on statistics, we obviously do also backups of the OS filesystem if necessary and wanted, but we did not encounter one failure in this decade on over 200 machines. All servers have at least two disks for the OS, some also 3 in mirrored configuration.
 
Last edited:
I was thinking of creating thread but there is no need now :),the reply i read at @DerDanilo post reply there is backup node so how to do that,I'm searching but there is no mention in proxmox docs.
then there is way to do that ?

Why don't you ask in the thread you read the reply instead of here? Chances of getting a useful answer are much higher, because all people that replied in that thread will get a notice that a reply is there. In this thread, probably only us three.
 
Why don't you ask in the thread you read the reply instead of here? Chances of getting a useful answer are much higher, because all people that replied in that thread will get a notice that a reply is there. In this thread, probably only us three.
Ok,thank you.
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!