Nginx Proxy Manager/Bridge Network MAC conflict

Apps and Virtualization
1

I just finished moving my network from a standard Netgear router to an Opnsense setup using a managed switch for hardwire connections. I’m currently not doing anything fancy with the switch, just letting it do its thing.

I had configured the NPM Scale app as a reverse proxy on my previous network and everything was working great. All of my other containers and the VM hosting Home Assistant were all accessible by the domains and subdomains I had set up.

Now that I have switched my network over, I am unable to access any of the services by domain name. I can, however, still access them all by their original IP addresses and ports that I had configured before.

My TrueNAS server is set up at IP x.x.x.44, and I had used Portainer to modify NPM to use my previously set up network Bridge and use the IP x.x.x.46. My Home Assistant VM is also using the network Bridge at x.x.x.37, and has its own unique MAC address.

The issue I think I’m seeing is that both the TrueNAS IP and NPM IP are using the same MAC address, even though both IPs are being routed correctly. I cannot find a way to change the MAC for either of these devices to something unique, like the HA VM is, which is what I’m assuming would allow NPM to start routing traffic correctly again.

One more thing, when I use the “ifconfig” command in shell, the only IP I get for the shared MAC address is the .46 IP. Although from my Windows computer on the network, “arp -a” shows both IPs with the same MAC.

Any insight would be helpful!

2

I’m not exactly sure how I fixed this - but I did find out that during the router swap, my new router swapped my TrueNAS static IP from .44 to .46 (as they were both listed as aliases under the bridge adapter), and after removing both IPs from the bridge, adding .44 (the intended IP) to my main adapter, and basically repeating the original process of moving that IP to the bridge, I was able to get the TrueNAS IP back to its intended value.

After some more fiddling, and more likely time, NPM began functioning correctly again and routing traffic correctly. I hope this helps someone with a similar issue in the future!