The initial ALP Announcement was this week, and the ALP Deployment instructions only cover how to test it on KVM. Since all of the support needed for Hyper-V is in upstream Linux, I thought I would give it a try. Technically this is all feasible, it's just a matter of making sure the right kernel options (or modules) are compiled in, and that the VM is configured appropriately
- Get the qcow image from ALP Download
- Download a qemu-img build for Windows from my friends at CloudBase
# Convert qcow image to VHDX
.\qemu-img.exe convert -O vhdx ALP-VM.x86_64-0.0.1-kvm-Build14.6.qcow2 ALP-VM.x86_64-0.0.1-kvm-Build14.6.vhdx
# Extend image to a larger size (40GB)
# Resize-VHD -SizeBytes 80Gb .\ALP-VM.x86_64-0.0.1-kvm-Build14.6.vhdx
# Create VM
$vm = new-vm -Generation 2 -Path . -Name ALP -BootDevice VHD -VHDPath .\ALP-VM.x86_64-0.0.1-kvm-Build14.6.vhdx -MemoryStartupBytes 4Gb
$vm | Set-VMProcessor -Count 4
$vm | Set-VMMemory -MaximumBytes 8Gb
$vm | Set-VMFirmware -SecureBootTemplate MicrosoftUEFICertificateAuthority
At this point, it's best to launch the VM in "Hyper-V Manager" so you can monitor boot process and complete setup.