Dear Tyge,
in case, you want to have multiple servers installed on one physical machine serving different domain names, you have at least the following options:
define for the same machine 2 IP addresses with the 2 domain names and run two NaviServer instances (two nsd)
use one IP address and register both domain names for it. Then define a subsite and register this via host-node map to the domain name (virtual servers by means of OpenACS). This way you have one nsd running, both sites share the same code base, using the same blueprint.
use one IP address and register both domain names for it. Then use NaviServer's virtual server to define two NaviServer servers in one nsd, and load OpenACS into both servers. This way you have one nsd process running, both sites have a different blueprint and can run different versions of OpenACS if wanted.
What you are doing above is the third variant, where probably the first two are simpler and sufficient. In most cases, the approach with the host-node map is sufficient and uses the least resources. We use this on our university for at least 6 domain names.
Hope this helps
-g
