VM's lose internet connection

Copper Contributor

I am experimenting with WAC, my setup is as follows:

 

1) HVHOST01 - i7 with 8 cores and 32gb ram running windows server 2022 core and the Hyper-V role

2) Host has 5 total NICs

3) Setup primary nic to 10.0.0.2 with gateway of 10.0.0.1 and DNS of 8.8.8.8

4) Setup virtual switch as a "team" for the other 4 nics

5) Deployed 4 Virtual Machines

   - Domain Controller

          * DHCP configured for 10.0.0.100 to 10.0.0.200 and successfully handing out addresses

          * DNS setup with forwarders configured as  8.8.8.8 and 8.8.4.4

   - SQL Server

          * Joined to domian

          * Assigned static IP of 10.0.0.50

          * Has internet access, however, after about 15 minutes the network connection will report No Internet Access

          * Only solution is to either reboot the server or disable/enable the virutal NIC

    - VMM Server

          * Joined to domian

          * Assigned static IP of 10.0.0.20

          * Has internet access, however, after about 15 minutes the network connection will report No Internet Access

          * Only solution is to either reboot the server or disable/enable the virutal NIC

   - Verified everytime this happens that the DC is up, all services are working and it still has access to the internet.

 

I have looked everywhere in WAC to see how I can adjust the NIC card properties of both the HOST and the VM's (i.e. VMQ) and can not find how to do this and have been unable to solve the problem of the network connection losing access to the internet.

 

Any help is greatly appreciated.

6 Replies
But when the "No internet access" message appears, you can still ping 8.8.8.8 for example? No internet access could also mean that DNS isn't working anymore or that the NCSI sites can't be resolved...
Did you have any luck in troubleshooting your issue?

@Harm_Veenstra , None.  Internal DNS seems to work just fine, but anything to the internet is "gone" as far as name resolution.

 

I'm at a loss as to why this is happening as it is only happening to the Virtual Machines, the physical host the VM's run on and of course the physical Domain Controller do not have this issue.

 

Thoughts?

 

Jason

Seen this in VMware with teamed networks and inconsistent switch configurations. The moment the issue occurs, all vm's on the host are inaccessible or just one? No port errors in the logs of the switches? Does the problem occur with a single nic for the host?
Seems to happen to ALL the VM's at once. There are 5 nics in the host and I've tried all 5 individually and in various team combinations and it happens regardless of that configuration.
5 nics as in one oneboard and a quad-nic? All the same chipset?