[one-users] Need help for VM template with a tricky network set-up

[Carsten.Friedrich at csiro.au]

I need to create a VM template which works in the following cloud / network setup:

I have an OpenNebula cloud where the cloud nodes are not all in the same subnet. To simplify discussions, let's assume I have 2 subnets (public IP address ranges) each hosting a cluster of OpenNebula cloud nodes. When I launch a VM I want OpenNebula to be able to chose any machine in any cluster to run the VM. Now, my VMs also get public IP addresses which have to be in the same subnet that the host is in. I can create 2 virtual networks one for each subnet and as long as I know which subnet the host is in I can chose the right network and everything works fine. However, if I want OpenNebula to be able to chose any host to run the VM, I won't know what subnet it ends up in until it gets deployed. Which also means I don't know which network I should put in the VM template as the template has to be created before the VM is deployed. Is there a way to specify a network conditionally on the cluster where the VM is going to be deployed? E.g. if the host  cloud node is in cluster 1, use network "red network", if it is in cluster 2, use "blue network"?

Cheers,
Carsten



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opennebula.org/pipermail/users-opennebula.org/attachments/20101014/4547de7d/attachment-0002.htm>