Differentiating between vm's

[cycloxr]

I noticed with proxmox backup that if you have say 5 different proxmox hosts, and each one of those hosts have a vm/100 and are backed up to the proxmox backup server that they all show up under the same vm/100 Backup group, is there a way to separate them based on the host they're being backed up from or to show the vm name instead of the id?

 

[dietmar]

The suggested solution is to use a separate datastore for each cluster (or host in your case). VM IDs are unique inside a cluster, so all
cluster members can use a single datastore.

 

[Dunuin]

I like to create a list with all VMID I'am using or I have used in the past. And each time I create a new VM I make sure to give that VM a new VMID and add it to the list. And if I need to restore/migrate a VM I make sure that the VM is created with the same VMID it previously had. That way no VMID will be used twice and you won't run into confusing situations where you backed up one VM as 100, deleted it from the host while it is still stored as 100 on the BPS, restored that VM as 101, backed it up as 101, so the same VM got VMID 100 and 101 on the BPS.

 

[cycloxr]

I would think just adding a hostname column would solve this issue, but doesn't look to be possible?

 

[suriv]

I would think just adding a hostname column would solve this issue, but doesn't look to be possible?

I don't think it's a good idea. What about HA? If VM will change the host which hostname should be added? This can produce a lot of confusion in future.

 

[toomanylogins]

I also found this aspect confusing. I was trying to use PBS as both a backup solution and an archive for some PVE guests where I'm saving an entire development environment. So I make a backup of a guest I'm no longer using and then I delete the VM from PVE. However creating new guests I now have backups which are out of sync with the VM ID. I'm currently using the comments column to try and keep track of this but it's confusing. Would appreciate any thoughts on the best way to manage this.

 

[antubis]

@suriv :

I think @cycloxr meant to add a column with the _VM_ (host)name?! IMHO this would be very helpful in the big forest of VMIDs - especially if you have multiple datastores for multiple PVE clusters/instances.

 

[VictorSTS]

Creating a relationship among VMs and the Host they are... run in? created at? looks bad to me.

IMHO, the proper approach would be to create an UUID (Universal Unique ID) when creating each VM/CT in Proxmox and use it to uniquely identify every VM worldwide. There are a few standards for UUIDs which might be applicable for this use case.

 

[oversite]

@suriv :

I think @cycloxr meant to add a column with the _VM_ (host)name?! IMHO this would be very helpful in the big forest of VMIDs - especially if you have multiple datastores for multiple PVE clusters/instances.good point

The host would not be the best column, in a normal cluster the host is not so important, VMs are supposed to be moved around. The cluster name however is a more correct column since VMids are within this realm. VictorSTS has a good point with his idea of UUID and as we are accustomed to that with drives, this leads me to think that it would be good to be able to choose. Cluster/VMid vs UUID or any other definition for PBS backup naming as you do with drives. One does not fit all. As long as VMid is so central in PVE, I believe Cluster/VMid would be a good default but leave it up to the user to use UUID for the backup naming.