Feature Request: Backup Restoration Jobs

D

deakus

New Member
Proxmox Subscriber
Jan 7, 2025
12
13
3
Screenshot_20250206_164639.png
Proxmox VE has the option to create/schedule backup jobs from within the Datacentre.

You can specify the VM's/CT's to be backed up and also what time.
It would be extremely useful to be able to do the same process but with Restorations

I can create something similar using the CLI and a bash script using vzdump

https://pve.proxmox.com/wiki/Backup_and_Restore

However I have not figured out a way when using Proxmox Backup Server.
 
  • Like
Reactions: Johannes S
If you are migrating to a new cluster. Or performing DR, Or performing a rip and replace to all the servers cluster.
You can connect your PBS to your new cluster. Create a new Restoration job of your vm's.
Run the restoration job, go back to the coffee shop and watch the progress on your phone.

Currently you have to point and click and search within the web GUI and manually restore the VMS one by one, or two by two to not swamp all your IOPS from your storage. As we have no way to set concurrent restorations you can set as you can like when performing migrations.

If we could perform scheduled restorations, we could even batch restore to avoid IOP storms.

Incidentally this the exact process i will be performing in 2 weeks time to a proxmox cluster
 
  • Like
Reactions: Kingneutron and Johannes S
alexskysilk said:
To add to what @bbgeek17 said, since this is not something most users would ever need, and the ones who do will need once in a blue moon, just script it. shouldnt be very difficult.
Click to expand...
You are right to a point. However lets take the view of wouldn't it be great if it was there out the box. Just like the fire extinguisher that you walk past every day when at work. Hopefully you will never need it, but its nice to know its there right !
 
  • Like
Reactions: Johannes S
Undoubtedly.

But consider what you're actually suggesting in context. In order to build this feature, someone (I assume you mean not you) would have to build, QA, test, bugfix, document, and support it. The devs have a ton of priorities both in terms of debugging, improving implementations, AND adding new features. Considering that a solution for your specific ask will take YOU 10 minutes to script, where would you put the developer resource spend in their priority list?
 
Its a feature request like any other, and hopefully be considered and chins scratched to see how useful it for others. I'm sure the devs went through the same process when considering to work on the DC Manager App. This might go on the roadmap, it might not, but it wont if no body mentions it :)

Consider this post
A

Thread 'Looking for advice on setting up disaster recovery/DR testing for a small company.'

I have a three server cluster that currently runs our primary application server and will be running additional servers over time. All servers are on the same rack/switch on 10Gbps ports + dedicated Corosync network via dedicated and isolated switch We have HA setup, Veeam running nightly backups, and PBS running hourly backups during business hours. So far so good, but I don't have the ability to test restore without actually restoring the production server.

Potential complication: Our application server is running from shared storage in the form of NvME media from a NAS also 10Gbps...

Others might be consider this feature useful, performing their DR testing might be part of an ISO 27001 standard and is part of their SOPS.
Having a well documented way of performing this task my prove beneficial.
Especially the amount of people looking at proxmox now as they are leaving VMWare behind.

Its just an ideas, not a demand
 
  • Like
Reactions: Kingneutron and Johannes S
deakus said:
Just like the fire extinguisher
Click to expand...
No workarounds for a fire extinguisher however (hopefully) infrequently it is used - a simple 10-min script isn't going to save you.
deakus said:
Its a feature request like any other
Click to expand...
Not exactly. Although it is not a totally useless feature, it is of limited use. The example post you gave of DR is actually a very complicated one & definitely needs a solution for anyone testing proper & correct server-wide/NW-wide DR testing.
 
I say it's a nice fetaure;). This week i make a script that backup the entire disk of a physical server with pbs-backup-client to a pbs server ,and another script with proxmox-backup-restore that restore the backup into a vm, on proxmoxve on a offsite site in case of a big disaster.
 
  • Like
Reactions: Kingneutron
alexskysilk said:
I dont have a useful reply, but could you share the usecase for this?
Click to expand...
One usecase would be to schedule reoccuring automated restore tests. This would be highly useful to ensure that backup and restore works as intended. Of course we have verify but an automated restore could also be combined with an automatic test whether a certain service and data can be accesses to ensure that no configuration error or something like that was made when creating the backup job in the first place (e.G. missing a data disc in the configuration)
 
Last edited:
  • Like
Reactions: Dark26 and Kingneutron
  • Like
Reactions: Johannes S and Kingneutron
You must log in or register to reply here.
Top Bottom