Hello everyone, I wonder if you can tell me how I can do this. I'm creating a VM image that needs to discover it's hostname, IP address information so I can configure /etc/sysconfig/network, /etc/sysconfig/network-scripts/ifcfg-[interface] and other things. If I know the oVirt/RHEV engine to connect to, I can get this information successfully. However, when creating a dummy 'gold' VM image and having oVirt clone it. I have no 'key' to tell me what engine to connect to, other than trying them all and hoping for best. I've written some code with the oVirt SDK that needs to know the URL: from ovirtsdk.api import API import platform import yaml import sys api = API(url="https://HOW_TO_DISCOVER_THIS/api", username="USER@internal", password="PASSWORD", insecure=True) ... Is there a way to know the engine to connect to when VM has booted up? I tried poking around ovirt/rhev-agent and noticed this: in /usr/share/rhev-agent/vdsAgentLogic.py: self.vio.write('host-name', { 'name' : self.dr.getMachineName() }) It *looks* like I could get the host name from what oVirt has configured w/o knowing the engine to connect to? But I cannot open /dev/virtio- ports/com.redhat.rhevm.vdsm (since rhev-agentd) has it opened 'too many files open' error. When using OpenNebula in past, they mounted a ISO CD image that contained the VM's metadata you could read in as configuration and having also used AWS EC2 You can connect to localhost to the VM itself to get metadata to get this sort of information. I really appreciate any help, if there's an easy way. Thanks, Shawn.