Hello,
I have a situation where we have two physically separate internet connections via the same ISP, one of which is connected via ether1 and the other is connected via ether2. The ether2 connection is faster but less reliable. I’ve attached a basic diagram of the layout.
We would like to use the ether1 connection for our servers (the machines that require static IPs and high availability) and for VPN and management connections to the routeros device, and use the ether2 for systems behind a NAT, with the NAT connection falling back to ether1 should the ether2 connection be disrupted. We’re not particularly interested in load balancing, just automatic fail-over. We would like to be able to connect to the routeros machine from either interface for management, VPN, etc.
Ether1 is part of a bridge, and some servers plug in there to get direct access to the line. These machines don’t need to worry about fail-over and must strictly communicate over ether1.
Issues we have run into:
- ether1 and ether2 are on the same subnet
- The two interfaces must never have the same MAC address
- ether2 will not route traffic originating from IPs assigned to ether1 (and vice-versa)
- Router must remain accessible from either of it’s two public IPs
I’ve tried following the instructions located here with disappointingly mixed results. Any assistance would be appreciated.
(EDIT: ISP was able to allocate one of our existing static IPs to the 2nd line)
