stress_junkie, thank you so much for your reply.
Almost but not quit. Sorry I have a very bad habit of not expressing myself very clearly...Will do my best
My clients connect to my server. I am using NAT in the VPN Subnet IP Pool. All is well as they connect to the internet through the VPN and they have access to the private subnets. All internet traffic is going through the server. What I dont understand is that they come out on the net with the VPN's server IP address. I had this crazy idea that the tcp header of the internal vpn addressing would be used internally and externally.
As per OpenVPN DOC, if you use NAT, each VPN client's virtual address is transformed via NAT so that the Access Server host's IP address is used as the source address on client packets destined for private subnets. Routing is so much more complicated but would use the virtual address of each VPN clients as the source address destined for private subnets. Using the later, routing of private subnets would need to be configured on the host so the response packets can be routed back to the VPN client via the access server host's IP address on the private subnet. Add this to the fact that I am using a virtual machine bridged to the host make my brain boils up when I tried to figure out routing. Some of us are blessed with inteligence and for the rest like me we need hard work
Finally, when the internet traffic if going through the VPN access server, the default route on a client seems to be pointing to the VPN gateway's virtual IP address.
Now, it does make more sense that the internet traffic from clients being filtered through the server, all the clients end up with the same IP which is the one from the Access Server.
Question: Is there a way I could manage each client to not have the Access Server Public IP address ?
Hope that clarifies a bit more..
Thanks
