Configuring multiple network connections on a single device can challenging. In this video, you’ll learn how to manage a wired and wireless network connection, manage IP routes, and discover neighbor devices.
<< Previous: Troubleshooting VLAN AssignmentsNext: Troubleshooting Mismatched MTUs >>
A very common hardware configuration these days is for our devices to have both a wired ethernet adapter and a wireless adapter in the same device. So the question is, which connection should we be using? If we plug-in the ethernet, should we still be using our wireless connection? And how do we make that determination?
The challenge, of course, is that your operating system is only going to use one interface for a particular route. So it’s going to choose either the wired connection or the wireless connection. And all of your traffic is going to flow through that particular interface.
Some operating systems will prioritize which it uses based on a setting in the operating system itself. It doesn’t use IP metrics or routing tables. It uses an internal determination in the operating system. Other operating systems might use the routing table. So they’ll determine what the best route will be and use that particular interface to send the traffic.
The challenge, of course, is the routing table may not be using the fastest connection. It’s simply using the best route that it happens to have in the table. So you may have to modify the routing table or delete or disable one of those interfaces to force all of the traffic through the other.
Here’s a routing table from my Windows device. We have two interfaces we’re most interested in. This top one is a Broadcom NetXtreme Gigabit Ethernet. And the other is a Broadcom 802.11n network adapter. These are the two that I’m using. In this case, I’ve connected to a wired ethernet, and I’ve connected to wireless simultaneously.
And if we look at my routing table, you can see I do have, indeed, two default routes. 0.0.0.0 with a netmask of zeros. You can see the gateway is exactly the same. Because my wired ethernet and my wireless network are on the same network. And they’ve been given two different IP addresses on the same subnet– 10.1.10.14 and 10.1.10.22.
And notice that the metric here is exactly the same. So it’s going to use the one that’s on the top, the .14, to send all of the traffic. But I want it to send all of the traffic on my ethernet connection, which is the .22.
Inside of the adapter configuration in Windows, there are some advanced settings that provide me access to the adapters and the bindings. And you can see the connections that I have on my device. And it does list the ethernet connection and the Wi-Fi 2, or my wireless connection.
If I look in my ethernet connection, I can see that it is using IP Version 4 in the client for Microsoft networks. But I might also want to look at the configuration of IP and how its metrics are configured. It is getting a DHCP address automatically.
But notice at the bottom of the screen I can change my interface metric to whatever I’d like it to be. And all I have to do is make sure that it’s smaller than 10 for it to take priority over the other interfaces. So in this case, I set my interface metric to 1.
If we look again at my routing table after making this change, you can see I still have the same interface adapters. They’re both still enabled inside of my computer. I still have those two default routes. But notice now that the metric has changed. 10.1.10.14 remains in a metric 10. But 10.1.10.22 now has a metric of 1, which means that all of my traffic out to the default route will always use the .22 instead of using the .14.
If you’re looking at your routing table and you notice that a route you were expecting to have in the routing table is now missing, then you’re going to have problems communicating to that particular subnet. We often see that when a link is missing or you lose a connection to your network and you have an outage. And that means you can no longer communicate to that particular subnet.
This is usually identified with a message. You can be at the command line. You can try pinging that subnet. And your operating system will tell you there’s no route to get to that host. We don’t even have a path. We can’t even tell you if the host is up or down. Because there’s no way to even get to that subnet.
Your pings or your communications may simply timeout, especially if the router itself is having a problem or it’s misconfigured with the incorrect routes inside of the router. Your packets will go into the router. The router will drop the information. And you’ll never get a response back.
If you’re trying to troubleshoot a route, you’ll need to look at every routing table between point A and point B. So you’ll want to see the routing table inside of your device. You’ll want to see the routing tables inside every router along the path. And you’ll want to see the routing table on the remote device, as well.
Another good check of network connectivity is to see if there’s anything local to you. There might be devices on your local subnet. And if you’re able to identify and communicate with those, that might help with the troubleshooting process.
Of course, you can always check your default gateway. Make sure you can ping that device. If the default gateway has been misconfigured or you have the wrong IP address, then you’re not going to be able to communicate outside of your local subnet.
So maybe you ping around to the local devices on your subnet. You can use third party devices like Nmap or IP scanners. Or you can simply use the Windows Networking function to see what devices it sees on the local subnet.
You can see I have different routers and network devices on my network. I have some network address storage. I have a Sonos device. I have a printer. Just by simply bringing up this Windows screen, it will automatically find these devices and show you that at least some of your network connectivity is working.