Virtual Datacenters

Virtual Datacenters is a concept for achieving user, group and resource isolation and at the same time allowing some form of software and hardware resource sharing.

All tasks created in Danube Cloud are executed inside a virtual datacenter. Especially the logging of all API functions is strictly related to the currently chosen datacenter, and there is a separate task log for each virtual datacenter.

Danube Cloud comes with a pre-installed public datacenter, which is available for all users. The datacenter name is main and it cannot be changed.

DC-unbound API functions

Some of the API modules and functions are not related to a chosen virtual datacenter and are always executed and logged in the main datacenter. This is especially true for compute node and user management related API functions.

DC-bound API functions

Most of the API modules and functions show, create, modify or delete virtual objects (virtual machines, disk images, networks ...) which are part of a virtual datacenter. These API functions are marked as DC-bound and always work in the scope of a virtual datacenter. Unless a datacenter is specified manually the selection of current datacenter is ambiguous and most likely to correspond to the datacenter chosen last via the GUI.

To specify a virtual datacenter, the resource must be prefixed with the /dc/(dc) path. Optionally, when using es, the current virtual datacenter can be set by using a -dc (dc) parameter.

es example:

es get /dc/admin/vm
{
    "url": "https://my.erigones.com/api/dc/admin/vm/",
    "status": 200,
    "method": "GET",
    "text": {
        "status": "SUCCESS",
        "result": [
            "mgmt01.local",
            "zabbix01.local"
        ],
        "task_id": "1e1d2-6f75849b-fc64-4e1c-9a7f"
    }
}

Note

Always remember to specify a virtual datacenter when using a DC-bound API function.