I have a following problem with static nat translation over vpn tunnel
My router is Cisco 1811 (let's call it central). We have a number of site-to-site VPNs between our offices. Those are set up with crypto ipsec to other cisco (871), linux distros, watchguards. Each office uses one of 172.16.x.0/24 subnets.
In here we have two WANs (which isn't that important - just giving a description to the config. there is no load balancing - rather a failover- which is not working not relevant now ), but our main one is 188.8.131.52. There is also a nat translation in use with route-map and pat translation in place.
- there is a customer we need to create a site-to-site vpn with. We should have access to their machine, they should have no access to our resources. We need to support that machine. Their remote WAN is 184.108.40.206. Their subnet is 192.168.1.0/24. The host we're interested in is 192.168.1.10.
I created a loopback interface with ip 10.168.168.33/32. Then created a VPN between this loopback and the remote host. It is working. From my local subnet (172.16.3.0/24) I can reach it using 192.168.1.10. Till this point - everything is working like a charm.
What I want and need to do is:
- assign a "virtual" IP 172.16.3.20 for example to do a translation from 172.16.3.20 to 192.168.1.10
Why? So that any of my users from 172.16.x.0/24 subnets can ping/rdp/call 172.16.3.20 and will be nated to 192.168.1.10. I don't know what is on the other end. Right now for the test environment I'm using Mikrotik device. I cannot ask the customer to add each 172.16.x.0/24 to their ACL. It has to be transparrent for them - in fact I don't know whether 192.168.1.10 is not just another proxy on their side. What's more, there will be soon another machine on their side (192.168.1.11) that we need an access to as well.
I tried various "ip nat inside source" and none worked. Can you help?