Hyper-V - NATAdapter & Network Adapter

[APT-410]

• Operating System Version: Windows 10 Professional
• Deploying via (VirtualBox/VMWare/AWS/Azure/ESXi): Hyper-V
• Vagrant Version (if applicable): 2.2.10

Description of the issue:

When building with the default switch option 1, during in the CMD prompt, all the switches use the NATswitch still at the end of the build instead of the default switch. I initially tested with option 2 NATSwitch and seemed to hang with WINRM in the build.

There is a second part as well, this is with a full build with option 1 for the default adapter. This error occurs after the shutdown of logger, after the machine is shutdown logger no longer has the IP 192.168.38.103 and is unable to communicate with the other machines, everything works prior to this. I tried to remove the NATAdapter and NATSwitch from logger, then in the GUI for hyper-v add the NATSwitch to the machine but doesn't seem to work.

IPs assigned immediately after deployment: 
PS C:\WINDOWS\system32> Get-VMNetworkAdapter -VMName *

Name            IsManagementOs VMName                SwitchName     MacAddress   Status IPAddresses                               
----            -------------- ------                ----------     ----------   ------ -----------                               
NATAdapter      False          wef.windomain.local   NATSwitch      00155DAB3524 {Ok}   {192.168.38.103}                          
Network Adapter False          logger                NATSwitch      00155DAB3528 {Ok}   {192.168.38.105, fe80::215:5dff:feab:3528}
NATAdapter      False          win10.windomain.local NATSwitch      00155DAB3526 {Ok}   {192.168.38.104}                                                              
NATAdapter      False          dc.windomain.local    NATSwitch      00155DAB3522 {Ok}   {192.168.38.102}  

IP assigned after reboot of logger:
Name            IsManagementOs VMName                SwitchName     MacAddress   Status IPAddresses     
----            -------------- ------                ----------     ----------   ------ -----------     
NATAdapter      False          wef.windomain.local   NATSwitch      00155DAB3524 {Ok}   {192.168.38.103}
Network Adapter False          logger                NATSwitch      00155DAB3528 {Ok}   {}              
NATAdapter      False          win10.windomain.local NATSwitch      00155DAB3526 {Ok}   {192.168.38.104}      
NATAdapter      False          dc.windomain.local    NATSwitch      00155DAB3522 {Ok}   {192.168.38.102}

### Link to Gist Containing Build Logs:
I didn't happen to see the build log generated, after build. I can add those generated during the build in CMD here tomorrow.

[clong]

Hi @apt-0 , unfortunately I don't support the HyperV provider for DetectionLab, but I'm going to tag in @man715 who developed the HyperV code just in case he has any thoughts!

@man715 - hopefully you don't mind me looping you into this issue!

[man715]

Hey @apt-0 ;

When building with the default switch option 1, during in the CMD prompt, all the switches use the NATswitch still at the end of the build instead of the default switch. I initially tested with option 2 NATSwitch and seemed to hang with WINRM in the build.

You will need to choose the first (default) option for the Network Adapter. That is how the VM is able to get an IP address and the host machine can communicate with the VM. If you try to have the VM use the NATSwitch from the start, it will not get assigned an IP address and Vagrant will not be able to communicate with it. After that, the scripts create a static IP for the box and use the NATSwitch that way the DetectionLab VMs are on their own isolated network.

There is a second part as well, this is with a full build with option 1 for the default adapter. This error occurs after the shutdown of logger, after the machine is shutdown logger no longer has the IP 192.168.38.103 and is unable to communicate with the other machines, everything works prior to this. I tried to remove the NATAdapter and NATSwitch from logger, then in the GUI for hyper-v add the NATSwitch to the machine but doesn't seem to work.

I am not sure why it would lose the IP address. I will try to replicate this on my end and get back with you.

[man715]

I have been able to replicate the issue of the IP address issue. I am working on fixing the cause. A workaround for now would be to edit the /etc/netplan/01-netcfg.yaml and remove anything after renderer: networkd. Then reboot the VM.

[APT-410]

Thanks alot @man715 ! That worked.

[clong]

@apt-0 are there any changes to the documentation I can add to help future users avoid this issue?

[APT-410]

Off @man715's suggestion, maybe just a note at end of hyper-v's readme known issues:

A workaround for logger losing network access/assigned IP or access to any of apps including splunk after a your first shutdown, would be to edit the /etc/netplan/01-netcfg.yaml on logger's machine and remove anything after renderer: networkd. Then reboot the VM.

To check if all your virtual machines have an IP assigned use in powershell: Get-VMNetworkAdapter -VMName *

---

[clong]

Updated the docs, marking this one as closed. Thanks for jumping in here @man715 !