I have an OpenNebula server with the following FS Datastore configuration:
- Files: 200GB (Host Storage)
- Storage1: 11TB (iSCSI SAN Mounted to Host)
- Storage2: 11TB (iSCSI SAN Mounted to Host)
Currently, OpenNebula is defaulting to “Files” datastore. When creating a Template’s storage, there is no option to target the disk’s datastore.
I’m seeking clarification if OpenNebula, will automatically provision VirtualMachine storage to “Storage1 and Storage2” once “Files” storage capacity is exceeded?
I’m not sure I understand your configuration. Are those Image or System datastores? Which drivers are they using?
In general terms, the storage works like this:
When the VM is deployed it needs storage for the images, and the system files and volatile disks.
- Images are created in a Datastore. When deployed, the images will (for most configurations) use that datastore storage. So if it runs out new VMs will not be deployed, other DS storage cannot be used.
- For the system files and volatile disks, the OpenNebula scheduler selects a System Datastore. If a system DS runs out of space, the scheduler will start to deploy new VMs in the other available system datastores.
More info here: http://docs.opennebula.org/4.14/administration/storage/sm.html