WilliamLam.com

How to Run Windows 8 on vSphere 5 (for reals)

by // 7 Comments

Awhile back I wrote an article about running Windows 8 on vSphere 5, but the trick did not work. There was an issue during the installation that caused a system panic in the OS and prevented Windows 8 to be installed. VMware did not officially support Windows 8 as a guestOS in vSphere 5 as Windows 8 itself was not released, only a developer preview was available.

It looks like VMware has been hard at work to get support for Windows 8 running on the vSphere 5 platform and I recently got a tidbit that the issue was resolved in a recent ESXi 5 patch. Make sure you keep this on the DL .... they maybe watching 🙂

Disclaimer: VMware still does not officially support Windows 8 on vSphere, this is mainly for educational and testing purposes. Do not use this in a production environment

Here are the five easy steps to run Windows 8 on ESXi 5

  1. Download and install ESXi500-201112001 (patch02) from VMware patch repository.
  2. Create either a Windows 7 or Windows 2008 R2 regular VM (must be either of these guests)
  3. Edit your VM and under "Hardware->Video Card" enable 3D graphics support (this is required if you plan on running VMware Tools). You will also need to use e1000 network driver, VMXNET3 does not work
  4. Install Windows 8 using ISO
  5. Sit back and wait for the installation to complete. 

Here are a few screenshots of Windows 8 running on ESXi 5

So there you have it, you can now run the developer preview of Windows 8 on vSphere 5!

FYI - VMware does have a KB article regarding Windows 8 on vSphere which I hear will be updated very shortly. Be sure to follow it for the full details when it is updated.

UPDATE: 

If you decide to install VMware Tools (which you should always), after a reboot, you may notice the system just sitting on the splash screen and not allowing you to login. The fix is to enable 3D graphics support. Power down your Windows 8 VM and edit the settings under Hardware->Video Graphics and check the 3D box and then power back on your VM.

Automating vCenter Server Appliance 5.0 (VCSA) Configurations

by // 2 Comments

I recently had a need to deploy half a dozen or so vCenter Server Appliance (VCSA) in my home lab and even though it is one of the easiest appliances to setup, going through the web management interface can still be time consuming. I was looking to see if there was a way to automate some of the configurations and while digging in some of the vCenter logs I came across /var/log/vmware/vpx/vpxd_cfg.log which provided the answer I was looking for.

If you watch the log as you configure the VCSA through the web management interface, you will see a series of commands calling /usr/sbin/vpxd_servicecfg. This utility is actually just a shell script wrapper for the configuration of vpxd (vCenter Server Daemon) and it's respective services.

As you can see there is a warning message about the use of the tool, so I will also re-iterate with my own disclaimer:

Disclaimer: This is for educational purposes only, this is not officially supported by VMware. Please test this in a development environment before using it on actual systems.

For my lab, this was exactly what I was looking for to quickly spin up VCSA and configure all the necessary services to start the vCenter Service. In my lab, I was able to get the VCSA booted up via DHCP and execute a simple shell script remotely via SSH to VCSA and then connect to vCenter Server after the process was completed.

Here is script with the minimal commands needed for running an embedded configuration:

#!/bin/bash

echo "Accepting EULA ..."
/usr/sbin/vpxd_servicecfg eula accept

echo "Configuring Embedded DB ..."
/usr/sbin/vpxd_servicecfg db write embedded

echo "Configuring SSO..."
/usr/sbin/vpxd_servicecfg sso write embedded

echo "Starting VCSA ..."
/usr/sbin/vpxd_servicecfg service start

To run the script remotely (you do not need to copy it to VCSA), use the following command:

# ssh root@[vcsa-ip] < configureVCSA.sh

The script accepts the EULA, configures the database as embedded configuration and then starts up the vCenter service. A return code is provided upon each command execution which has VC_CFG_RESULT=N where N can be 0 which is a successful execution and any other value means there is an error. To figure out what the code means, there is a section in the shell script that lists all the codes and their meaning. In general, you should be getting 0 for all commands

Here are some other useful commands that I tested with:

Configure Active Directory:

/usr/sbin/vpxd_servicecfg ad write [AD_USERNAME] [AD_PASSWORD] [AD_DOMAIN]

Re-initialize vCenter DB *** PLEASE BE VERY CAREFUL WHEN USING THIS COMMAND, AS DB IS WIPED***

ensure vpxd has stopped

vpxd_servicecfg stop

re-init DB

vpxd_servicecfg db initialize

You can also configure an external database for vCenter, NIS directory service and even replace SSL certificates but I will leave that to the reader for some exploration.

Unattended Deployment of vCenter Infrastructure Navigator

by // 2 Comments

I deployed VMware's new vCenter Infrastructure Navigator in my lab over the weekend and just like the rest of the other virtual appliances (vCloud, vCO, vCC, vShield), here is how you can automate the deployment of VMware vIN.

Here are the ovf parameters that are available to deploy vCenter Infrastructure Navigator:

  • vm.password
  • vami.gateway.vCenter_Infrastructure_Navigator
  • vami.DNS.vCenter_Infrastructure_Navigator
  • vami.ip0.vCenter_Infrastructure_Navigator
  • vami.netmask0.vCenter_Infrastructure_Navigator

To see these properties before deploying, you can query using the ovftool which can help you identify the name of the ovf variables using the following command:

ovftool --hideEula Navigator-1.0.0.49-592384_OVF10.ova

Note: Before deploying vIN, ensure that you have the vCenter advanced setting VirtualCenter.ManagedIP configured as it is needed by the vService in vIN. For more details, take a look at this blog post on how you can easily automate this.

Here is an example of the ovftool command to deploy vIN Server:

ovftool --acceptAllEulas --skipManifestCheck '--net:Network 1=VM_Network' --datastore=iSCSI-4 --diskMode=thin --name=vin --prop:vami.DNS.vCenter_Infrastructure_Navigator=172.30.0.100 --prop:vami.gateway.vCenter_Infrastructure_Navigator=172.30.0.1 --prop:vami.ip0.vCenter_Infrastructure_Navigator=172.30.0.150 --prop:vami.netmask0.vCenter_Infrastructure_Navigator=255.255.255.0 --prop:vm.password=vmware123 Navigator-1.0.0.49-592384_OVF10.ova 'vi://root:*protected email*/?dns=vesxi50-3.primp-industries.com'

Of course, I wrote a simple shell script deployvIN.sh to help with the deployment. The script assumes you have ovftool installed and the OVF files located in the same directory as the script. You will need to edit the following variables if you wish to deploy vIN:

Note: There are many ways of using the ovftool to deploy an OVF. In this simple example, it requires you to specify an ESX(i) host, but you can modify the locator to deploy to a VM folder or datacenter path. For more examples and options, please take a look at the ovftool documentation.

Here is an example of the script in action:

Once the vIN virtual appliance has been deployed, you can also have it automatically power on by specifying the following parameter --powerOn.

If everything was successful, you should be able to license vCenter Infrastructure Navigator using the vSphere Client C# client and then login to the vSphere Web Client to enable the discovery process for your virtual machines. Shortly after, you should start seeing some application dependency within your vSphere environment like this: