I would like to use ISP2 instead of ISP1 internet from PC1 computer. But I'm not even able to ping antenna A, however, antenna B shows that it is connected to antenna A. I would also like to access all devices on 192.168.1.x subnet and vice versa. What I'am missing?

Antenna A and Antenna B configuration


  • Perhaps draw a small diagram and edit your question to include it. ISP 1 and 2 will have different external IP addresses and so a computer on the inside of this would need an arrangement like VPN to connect to the other subnet. I see your scheme now. You need something to connect the external IP addresses – John Feb 15 at 18:35
  • @John I'm using two Ubiquiti antennas (as shown in diagram) to connect two separate LANs. Both antennas connects to each other, but I still can't access 192.168.1.x subnet from 192.168.0.x subnet. – Luke Feb 15 at 18:51
  • I understand the diagram, but you cannot just connect across two different external IP addresses supplying each wireless. You need to connect the external IP segments. – John Feb 15 at 18:56

Your devices do not know that the other subnet is directly bridged. For example, PC 1's address & netmask tell it that only 192.168.0.x is on-link – anything else is to be sent towards the gateway. Then the exact same thing happens on both routers C and B. (Eventually the packet is sent to ISP1 and discarded there.)

There are several ways to connect the two subnets:

One large subnet

If you change all netmasks to /23 (i.e., you will end up with all devices being in one subnet and they'll know that they can reach each other without using a gateway.

When doing this, you should enable the WDS Bridge option, since it preserves MAC addresses through the bridge. (Standard Wi-Fi cannot do that, so without the WDS option the entire network B will appear to have a single MAC address when looking from network A.)


You can keep the netmasks as-is, and give certain devices IP addresses from both networks – for example, PC1 can be and at the same time. (Unfortunately on Windows this is only possible with static configuration, not with DHCP.)


Instead of transparent bridging, switch one of the antennas for 'router' mode. (Do not activate NAT – just routing.) Configure it with an 192.168.0.x address on one side and 192.168.1.x on the other.

Then configure router A to have a static route for via the antenna's 192.168.1.x address, and do the exact opposite on router C.

(Note this doesn't need to be done on the antennas, it could be handled by a separate router as well. For example, Router C itself might be capable of this, as long as the OpenWRT doesn't insist on "LAN vs WAN" and just lets you configure two coequal networks.)

  • It works! I've used "One large subnet" approach, because my router A, does not support static routing. Thank you very much! – Luke Feb 16 at 12:50
  • Hmm, there's a big issue I completely forgot to mention: if you use bridging and your both routers have DHCP enabled, you will often have issues with devices getting an address from the 'wrong' side. (Only the latency due to distance currently prevents that most of the time.) So better make sure all devices have 'static' leases... – user1686 Feb 16 at 14:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.