Trying to understand the Storage "consumption model"

Community Support Storage
1

Hi.

New to OpenNebula coming from the VMWare stack I’m struggling with mapping concepts from there to OpenNebula. Maybe not possible?

My setup is:

  • Frontend running outside my KVM-cluster as VM with no connection to SAN
  • KVM Hosts using SAN (iSCSI) and LVM - using BRIDGE_LIST to configure

I simple want to create a few persistent VM’s with persistent storage.

Vm1:

  • Disk1: 60GB - Rocky8 installation
  • Disk2: 600GB block storage where the app will partition that disk so no FS from OpenNebula.

Vm2:

  • Disk1: 60GB - Rocky8 installation
  • Disk2: 600GB block storage where the app will partition that disk so no FS from OpenNebula.

I’m struggling with if that extra disk (Disk2) should be an IMAGE disk or something else?
Created a IMAGE disk with size 600GB and type=DATABLOCK. For Image location, I went with “Empty disk image” hoping that it would not be created on the Fontend. But unfortunately, it does.

Why would an empty disk be SSH copied over to the KVM host?
We are not using NFS.

Am I going about this the wrong way?
Please help me out. Would really appreciate it!