Restoring a Backup to all original drives on the VM?

CyborgRider

New Member
Feb 25, 2024
7
1
3
Okay, This has been a recurring problem for me. I have multiple drives attached to my trueNAS vm. I back these drives up to my PBS on the same network, and this works! I'm able to backup the whole 4.7TB of data without issue. However the issue occurs when I begin to Restore to the node. I have had to rebuild the node, and when I did, I reset all the drives to the exact same naming scheme. I then hit restore backup. I expected the data from the OS to go to the disk I had the OS on, the data to go to the disk I had the data on, and so forth. But of course it didn't. Instead, I get an error saying just one of my disks isn't big enough to handle the entirety of the drives connected to the VM. I'm aware! I used several disks to achieve this setup, and when I restore it, I want it back exactly how it was backed up!

So my question can be simplified to this: How do I restore a backup from PBS, and have all of my drives go back in place, instead of trying to all fit on one of my disks and returning a frustrating error? What can I do to make the backup process more effective?

For now, I can pull the large files off my NAS one by one with file restore, but that downloads to my current PC instead of the VM that I need them back on. Am I using PBS wrong, or is it just this limited? Can we get an update that restores all drives to their original Disks? What is the next step for a situation like mine? I don't see much on google about it, so I haven't found a good workaround. I searched the forums, but I didn't see much here either. Hopefully someone has seen this before.
 
I'm asking now, for my personal use, but I will likely have the same issue with PBS on my work servers that have Enterprise licenses.
 
I think for the moment, the product is acting as expected.
I saw a feature request on their bugzilla to allow you to select which disks restore where.
I just tried to find it for you, and I can't. Their bugzilla is ... um ... well there's a lot of stuff there.

I ran into another user with a similar scenario, except he was using an lcx container, not a VM.
LCX backups let you exclude directories.
For that person, I recommended they do several different backup jobs, one for each drive.
The restore process would be ugly tho ... They didn't like that idea. Not sure what their eventual answer was.

Here's another idea. Break out these datastores into discrete NAS instances.
Clone a couple more NAS VMs, one for each physical storage.
Of course, you'd never do that with a hardware NAS, but then ... you'd never have this problem with a hardware NAS.
 
Ah. It was a thread.
Here. Read this.


Make sure to read the last post. Fiona is a dev, works on storage and backup. It's the word of God. (Well, if God had to have everything signed in triplicate and pass code review before dispensing miracles.)
 
Last edited: