Using Hyper-V as a Hobbyist’s Hypervisor (Updated)

Microsoft logoHyper-V is a bit of an oddball in this set, if only because of its ancestry. When you read the description at the Microsoft site, it becomes clear that you’re not getting just a Hypervisor, you’re getting Windows Server 2012… JeOS‘d. (pronounce as “Juiced”, from “Just enough Operating System”) Hyper-V is officially just enough of Windows Server 2012 (previously known as Windows 8 Server) to run the Hypervisor, but that does include quite a bit, because it has to allow remote management through the Microsoft Management Framework. But let me put this up front: of all the boot animations, this one is definitely the coolest so far. 🙂

Installation of Hyper-V is just like, well, installing a recent Windows. Predictable and simple. The result is a host that boots into a Windows 2012 logon screen, where you press Control-Alt-Delete to log on as “Administrator”. Only then do you notice the major difference (if you’ve never played with Server “Core” installations before) in that you just get a CMD window. Well, two actually; one normal one and a second that runs a configuration script. The configuration script allows you to configure networking, enable remote management, set the time, and do the basic shutdown/reboot stuff. The other window allows you to do pretty much anything you want in Windows Server land, which nowadays is running Powershell. For Windows Server, if you want to go commandline, Powershell is the way to go, but for this exercise I want a GUI.

Installing the Management GUI

Ok, here is the major issue I have with Hyper-V: it is built to be managed in a Windows Domain, using a recent Windows box. Before you search for the comment box; yes, you can do it without; I did. It’s just not as trivial.  But even then, if you don’t have Windows 7 or 8 in the taste Professional, Enterprise, or Ultimate, you are out of options. I cannot really blame Microsoft for this, but I am disappointed.

I started with just the documentation directly available and got stuck. I had to do lots of tweaks on both server and client to get connectivity, but even then I could not get the advertised “Server Manager” to work. This apparently was the result of a lack of an Active Directory domain and missing management frameworks. Luckily Google is my friend, so I found this article, which gave a good step-by-step. Initially I was surprised it put so much emphasis on enabling remote management of Windows Firewall, but when I wanted to use Disk Management I found out why: all those “area managers” have their own firewall rules that need to be enabled individually. You don’t say “I want to manage this system remotely,” but rather “I want to do Xxx Management remotely.”

Managers and Snap-ins

My “Administrative Tools” folder has doubled in size. I had initially installed the Server Manager as well as several other management tools, unsure what I would need. In the end, I only used two things: “Computer Management”, which is what you can open by right-clicking “Computer” and choosing “Manage”, and the Hyper-V Manager snap-in to the Microsoft Management Console utility. The latter is actually a generic GUI which you fill with snap-ins, after which you can save the state (the choice of snap-ins) under a new name in the “Administrative Tools” menu. The Hyper-V Manager is where you do your Hypervisor management, and it works great. It even has a thumbnail-sized image of the console, but if you connect to the VM, you get a Remote Desktop Client session. Kind of funny, that: an RDP session showing a Linux screen. 🙂

Installing and the Passthrough Tests

Setting up a VM and installing Ubuntu was a breeze. As I set, the VM Manager works great. Ubuntu had no complaints at all and worked fine. As for the passthrough tests, I can be quick about them: Neither my simple SATA disks, nor the network card, were available for physical passthrough. Not Enterprisey enough. I actually found a post on a Microsoft forum that had a Microsoft Support person answer a similar request with a curt list of supported hardware; all too high class for my system it appears.

Update: I talked to one of my colleagues who knows a lot more about Windows Server 2012 and he assured me the disks should not be a problem. Looking back at all posts on adding Physical Disks I did notice they all used the Server Manager, which I never got to work on my Windows 7 Ultimate box. I downloaded a Windows Server 2012 Standard Virtual Image from the Microsoft site and ran that using VirtualBox. Using this I was able to manage the disks and take them offline in such a way that they became available for use in the Client. I also found I could view the disks using PowerShell and do the online/offline switching there.

 Conclusions

Let’s see what we have in the Good vs Bad department.

Good

  • Hyper-V Server 2012 works great.
  • Hyper-V Server installation is quick and painless.
  • Management GUI, if you get it working, is great.

Bad

  • You need a recent, business or ultimate type Microsoft OS for the Admin GUI.
  • No passthrough, so all your disks need to become virtual disks, unless:
    1. You have a Windows 8 or Server 2012 system to run the Server Manager, or
    2. Manipulate the disks from the shell.

This one was fun, but I don’t think I’ll use it.

Advertisements
This entry was posted in Hands-on, Tools and tagged , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s