Example of how to use Set-VMFirmware to add an EFI shim path (VMBootSource) ?

[atsaloli]

Prerequisites

• Existing Issue: Search the existing issues for this repository. If there is an issue that fits your needs do not file a new one. Subscribe, react, or comment on that issue instead.
• Descriptive Title: Write the title for this issue as a short synopsis. If possible, provide context. For example, "Typo in Get-Foo cmdlet" instead of "Typo."
• Verify Version: If there is a mismatch between documentation and the behavior on your system, ensure that the version you are using is the same as the documentation. Check this box if they match or the issue you are reporting is not version specific.

Links

https://github.com/MicrosoftDocs/windows-powershell-docs/blob/8a35f5e26886d3b04687eca3ecea23d91943d6a9/docset/winserver2022-ps/hyper-v/Set-VMFirmware.md

Summary

Could you please add an example showing how to use Set-VMFirmware to add a VMBootSource firmware path?

Details

For example, I have a bootable Gen2 (UEFI) VM where I installed Debian that has the firmware path pointing to the debian shimx64.efi:

PS C:\Windows\system32> (Get-VMFirmware -VMName gen2-debian-test).BootOrder | Where-Object { $_.BootType -eq 'File' } | Select-Object -ExpandProperty FirmwarePath
HD(1,GPT,FFB61BB9-ADBE-49BC-BCE8-22AF4F699210,0x800,0x100000)/\EFI\debian\shimx64.efi
PS C:\Windows\system32>

I want to add a similar setting to another new Gen2 VM, but I am having trouble figuring out how to do that.

I know the disk identifier, but I just can't figure out how to set the firmware path using Set-VMFirmware.

Suggested Fix

Add an example showing how to set firmware path to EFI shim, please.

[websafe]

+1

[luckylinux]

@atsaloli : same Problem at work on a Windows 11 Workstation Laptop with HyperV Manager.

I was trying to use these Powershell Commands to basically "clone" a Template Archive containing a Folder Structure generated from Export-VM (and .vhdx Hard Disk, .vmcx Configuration, etc).

I also was pulling my Hair out to be honest for a few Hours 😕 .

To be fair, at least for MY Case/Issue based on MY Experience (i.e. Importing a Virtual Hard Drive Image using Add-VMHardDiskDrive into the newly created VM that was setup using New-VM), it's actually NOT needed to add this VMBootSource Entry by Hand.

Maybe it's just not possible. Or it's automatically detected by HyperV Manager. Or these NVRam Objects are setup during grub-install or update-grub. Not sure why it just automatically detected it after 1 Startup (and possibly 1 Shutdown).

I also attempted to copy the Object that was present in the .vmcx Configuration File (you need to analyze it using Compare-VM -Copy -Path if you want to go down that Rabbit Hole) but of course it has the wrong $VM.Id and $VM.Name associated with it. And trying to modify those yields a Read Only Property Error.

What worked (nothing special really):

1. Create a new VM using New-VM
2. Import the Virtual Hard Disk with Add-VMHardDiskDrive
3. Set the newly Imported VM Hard Disk as the (sole) Boot Device

(I just setup the VM Hard Drive as the only Boot Device for the VM)

As soon as I booted the newly created System (Ubuntu 24.04 AMD64 VM), the File Type Bootentry was automatically created/detected pointing to the same shimx64.efi File. No boot Issues at all (I was actually surprised by this). Strangely the Template had 2 File as VMBootSource pointing to 2 different shimx64.efi Files / Partition IDs. The New VM only had 1. Well, it seems to work anyways 🤣 .

I know this is not really answering your Question (or mine) 😞. But this is what I could observe at least. Secure Boot was ENABLED as per the Source/Template VM. Both Template and New VM are of Generation 2.

Anyways, it seems to "just work" (knock on Wood), at least like this.