Difference between site-to-site and client-to-site configuration on openVPN server

Question

I am trying to setup a site to site VPN tunnel using openVPN. The server currently is setup to accept client to site connections.

When configuring a site to site tunnel (as opposed to the client to site that is currently setup), is there something different that needs to be done in order to have both client to site and site to site?

I suppose my question is: from the servers point of view, is the configuration the same, regardless of whether there will be a site-to-site connection or a client-site connection.

Answer 1

From OpenVPN's perspective there's no difference between a site-to-site and client-to-site connection, it's just a matter of the configuration you want. You need to decide whether you can support both with a single server instance, or whether you need two server instances running on different ports.

A site-to-site VPN will typically have iroute directives to route to subnets behind the client, whereas a client-to-site VPN won't. Since iroute directives are per-client this can be done in a single instance. You may want to push different routes however, or use a different VPN subnet, or push redirect-gateway, in which case you probably need a separate instance.