How do people manage NAT with nebula?

libvirt allows for per vm (per libvirt host) nat rule generation via
the (debian) hook "/etc/libvirt/hooks/qemu"

However, nebula provides primary key domain names (on the libvirt
host) rather than using the machine name (viz. one-47 vs. oneiric).
Hence, to have machine specific NAT rules brought up and down with the
machine nebula would need to write to a database (text file) that the
hook script accesses.

e.g. as a kludge, presumably there is a pre-image-transfer nebula hook
that I could call and just do something like

echo "machine details" > file; scp file
libvirt_host:/etc/libvirt/hooks/machines/.

where the nebula hook would know the details like machine name and
also the targeted host

Does something like this already exist? Is there a better "best
practices" for managing per machine NAT rules with nebula?
_______________________________________________
Users mailing list
[email protected]
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org

Reply via email to