Windows 11 Multiple Gateways appended to DHCP ethernet adapter

Question

I am experiencing an issue with our Windows 11 24H2 running on vSphere 8. We have a usecase where we call the vSphere API to change the portgroup associated with a secondary NIC on the windows 11 desktop virtual machine. When this happens the ethernet adapter pulls a new DHCP address and gateway however it is not deleting the old networks gateway instead it just appends it. When we switch it back to the original network or select another network it just keeps appending sometimes we have as many as 6 gateways. The only way is to manually reset the ethernet adapter, but this isnt a viable solution for our usecase. This behavior does NOT happen in Windows 10.

Answer 1

Hello Anthony M,

What you are seeing is a change in behavior introduced with Windows 11 24H2’s networking stack. When the NIC is reassigned to a new portgroup in vSphere, the adapter correctly pulls a new DHCP lease, but instead of replacing the existing default gateway, the stack appends it. This is why you end up with multiple gateways in the routing table. Windows 10 did not exhibit this because its DHCP client logic cleared the previous gateway when a new lease was obtained.

The underlying cause is that Windows 11 24H2 uses the Next Generation TCP/IP stack with updated DHCP and NCSI (Network Connectivity Status Indicator) behavior. When the NIC GUID remains constant but the network changes, the DHCP client treats the new lease as an additional configuration rather than a replacement. This is especially visible in virtual environments where portgroup changes are frequent.

To confirm, run route print after switching portgroups. You will see multiple 0.0.0.0 default routes with different gateways. The only way they are cleared is by resetting the adapter, which forces a full reinitialization of the interface.

At present, there is no registry or Group Policy setting to revert to the Windows 10 behavior. The supported workaround is to script a reset of the NIC after portgroup reassignment. For example:

netsh interface set interface "Ethernet 2" admin=disable

netsh interface set interface "Ethernet 2" admin=enable

This clears the stale gateways automatically. Alternatively, you can use PowerShell’s Restart-NetAdapter -Name "Ethernet 2" in your automation sequence after the vSphere API call.

Microsoft is aware of this regression in 24H2, and it is being tracked internally. Until an update is released, the only viable solution is to integrate NIC reset into your workflow.

I hope you've found something useful here. If it helps you get more insight into the issue, it's appreciated to accept the answer. Should you have more questions, feel free to leave a message. Have a nice day!

Domic Vo.