Quick Fix: Virtual Machine Could Not Be Started Because The Hypervisor Is Not Running

I got smacked with this error message today which says the virtual machine could not be started because the hypervisor is not running as seen below:

I double checked Intel-VTx and DEP was enabled in the BIOS and then discovered there were no event logs under Application and Services Logs > Microsoft > Windows > Hyper-V-Hypervisor in the Event Viewer. I would’ve thought an event or two should be logged since I enabled the hyper-v feature and also created a virtual machine.

Fortunately there’s a short one-liner I ran in an elevated command prompt to set the hypervisor to launch upon boot up:

bcdedit /set HypervisorLaunchType auto

You’d think the hypervisor would be set to launch automatically after having enabled the feature in the first place, right?

My experience with Hyper-V up to now has been on the server side with the fully fledged hypervisor on Windows Server 2012 R2. So I decided to make use of client hyper-v on my Windows 10 computer for building reference images when I hit this snag

My Lab at Home

I’ve always felt there was something about the Lab that didn’t sit quite right with me which held me back from writing this post sooner. Investing in a HP MicroServer recently was a good move and having it set up as my vSphere host has definitely set things right in my eyes. I quite like my current set-up now.

I’ve mentioned my lab quite a few times on my blog but this is the first time I go into detail on its set-up. So here it goes…

I have VMware vSphere 5.0 running on my vSphere host as my hypervisor which currently has two guest Virtual Machines – Windows Server 2008 R2 and Ubuntu 11.10. I have an Active Directory domain set up on the Windows Server along with a Windows 7 deployment environment with MDT 2012 and Windows Deployment Services. On Ubuntu I have a FOG server (a Linux-based cloning solution) up and running which is mainly used for disaster recovery and backups (useful for image building in deployment scenarios as well as for personal use).

With regards to physical machines I have a Windows 8 PC which I use to administer the vSphere host using vSphere Client and for connecting to the Windows Server and Ubuntu VMs using Remote Desktop. There are also two spare computers in the lab for testing purposes, either for Windows 7 deployment or as client PC’s in my Active Directory domain.

vSphere Client 5.0 for Windows 8

I upgraded one of my computers to Windows 8 today. It seems the Windows 8 Upgrade Assistant had falsely led me to believe that vSphere Client 5.0 was compatible with the new OS, when in fact it wasn’t.

Try as I might, I just couldn’t get it to install and compatibility mode was of no help either. I tried a few other things too but let’s cut the chase and get to the point here.

In my search for a solution I came across a blog post with direct links to a number of vSphere Client installation files. I can confirm the vSphere Client v5.0 Update 2 is compatible with Windows 8.

UPDATE: This is also works with Windows 8.1 RTM

UPDATE: This is also works on WIndows 10 1507, 1511 and 1607

Installing VMware vSphere 5.0 – Part 2

In this follow up to Part 1 I will take a look at configuring the network settings on the vSphere host.

Once you’re through with these instructions there’s no need to have it connected to a monitor or keyboard as you can use the VMware vSphere Client to administer the host remotely.

Turn on your host and wait for the hypervisor to load. How long this takes will depend on your hardware. (It takes a couple of minutes on my HP MicroServer N40L with only a dual core CPU.)

Press F2 on your keyboard to get to the configuration pages of the host

Type in the root password you created at the time of installing the hypervisor

Installing VMware vSphere 5.0 – Part 1

The instructions here are pretty straightforward so I’m going to keep it really simple.

Note: My first attempt was to use a Virtual Machine to boot off the ISO (without having to burn it to CD) but my installation hanged midway. I had to burn it to a CD in the end. Personally I installed my hypervisor on a USB drive using a different machine than my microserver since the server obviously doesn’t come with a CD drive.

Once you boot into the installation CD this is what the boot menu looks like.

Upgrading the HP MicroServer

I’ve just finished upgrading the server with 8GB RAM and a 1TB hard drive and I’ve taken plenty of photos for step by step instructions on how to go about upgrading it.

Unlocking the door on the front of the machine reveals the inside where there’s four hard drive caddies in a cage and the motherboard tray screwed onto the base with a couple of hex screws. There’s a small allen key mounted on the door on the inside – this is shown below marked with an arrow.

First stop is the hard disk, which is mounted on the first caddy from the left. Press down where shown on the picture below and lift and pull the lever to release the hard drive.

You will need to use the allen key to unscrew the hard drive to release it from the caddy and fix your own hard drive onto it. Once that’s done keep it aside and remove the remaining three empty caddies from the cage to give us some room to get to the motherboard.

Introducing the HP MicroServer N40L

I’ve just taken delivery of the MicroServer which I bought from eBuyer a few days ago. It was £220 upfront but HP is offering a £100 cash back – great value for £120.

It’s a 1.5GHz AMD dual core processor with 2GB RAM and 250GB hard disk. I’m going to boost the RAM to 8GB along with a 1TB hard drive which I’m expecting to arrive in a couple of days.

The maximum RAM the server supports is allegedly 8GB but some people have managed to get 16GB memory (2x 8GB) working on this. Though its definitely a hit and miss with the RAM, the 8GB RAM that I ordered is one of those sticks that people have reportedly paired to get 16GB working on this server. This keeps my options open if I want to try my hand at boosting the RAM to 16GB myself. For now I’m sticking with 8GB – let’s see how this works out for what I want to do.

Speaking of which, the intention is to learn about virtualization, specifically running virtual machines in VMware vSphere 5. Right now I’m thinking of running Windows Server 2008 R2 and migrating my physical Ubuntu box, which runs my FOG server, to vSphere 5. I’ve already downloaded vSphere 5 along with VMware Converter Standalone which I’ll need for the P2V migration.

I’ll come up with a post on upgrading the server once the RAM and hard drive is here.

UPDATE: Just got an email from HP accepting my cash back claim. The cheque should arrive within 45 days.

Getting VM’s to Play Nice With Fog

I had an interesting experience trying to get virtual machines in VMWare Player to work with FOG – there was a trial and error process trying to get the VMs to PXE-boot into FOG and image the hard disk, etc.

This is a quick post just to note down some pointers to bear in mind when attempting the above:

• Use ‘bridged’ network type
• Use a static MAC address
• Change the disk type to IDE
• Allocate all disk space to begin with
• Update the fog kernel image (I used 2.2.4)

I’m coming up with a companion video to demonstrate the above on YouTube soon – It’s about time I dug out my screen capture software, Snagit, somewhere from my backups.

UPDATE: The video is now up on YouTube