Someone just pointed me to the new VMWare vSphere 4.1 key features and comparison chart.  They’ve put together a document that is supposed to compare vSphere 4.1 Enterprise Plus with Hyper-V R2, Citrix XenServer 5.5 and RHEV 2.2.  I always find it amusing how VMWare compares their most expensive SKU suite to everyone else’s stand-alone hypervisors…but I digress…

Notice in this example how the Enterprise Plus get's the check mark as if this new ‘large scale management’ feature is included with that SKU.  If you didn’t know, Virtual Center does not come with Enterprise Plus – far from it.  This conveniently ‘included’ feature happens to cost $7,318.00 (w/ 2yr maintenance).  That’s IN ADDITION to the vSphere Enterprise Plus license:

Which begs the question - Why doesn’t the VMWare box say “Requires vCenter Server” like the Hyper-V box says “Requires SCVMM”?  Instead of doing that – they stick it in the key feature column and check the box for Enterprise Plus.  Sneaky. 

I could go on, but my point of this diatribe really has little do with VMware or any other vendor for that matter – this just presented a good example for me to make my point - it’s not just about the features.  Every vendor on this list has features and capabilities that the other vendors don’t have and frankly any vendor can and has ‘creatively marketed’ something like this to you.  If you are involved in a virtualization project it’s imperative that you first understand the workloads that will be virtualized and what capabilities that you require prior to immersing yourself in the marketing and making decisions based on whose list has the most checkmarks.

If you are a partner or customer involved in a virtualization project make sure you reach out to someone at Microsoft – we can help you.

So, I logged into my lab this morning to once again find that BOTH of my Hyper-V server’s had restarted because of a blue screen issue.

This is what I saw:

Problem signature:
  Problem Event Name: BlueScreen
  OS Version:      6.1.7600.2.0.0.274.10
  Locale ID:           1033

Additional information about the problem:
  BCCode:             101
  BCP1:  000000000000000D
  BCP2:  0000000000000000
  BCP3:  FFFFF880022E2180
  BCP4:  000000000000000C
  OS Version:      6_1_7600
  Service Pack:    0_0
  Product:             274_3

These are HP Server’s running dual Intel Xeon 5520 Nehalem CPU’s.

Ah, what to do…

• I download the x64 Debugger Tool and install that.  I need to see what the dump says about this issue…
• After the install, click START and type “windbg” and run that application
• The first thing you’ll want to do there is point the debugger to the symbols path by using FILE –> SYMBOL FILE PATH.  Use this URL: http://msdl.microsoft.com/download/symbols
• After that, you get to open the minidump file.  Windows typically sticks them in the \windows\minidump folder unless you change it.
• Click FILE –> OPEN CRASH DUMP
• Browse to the location of the dump file you want to examine and open
• It will take a minute for it to complete, be patient. :)

Now, when I first ran this on my server – it came up with this:

Make sure you click the link to ANALYZE –V to get detailed debugging information.  In my case, this came up:

So, now I know the issue.

The Fix:

http://support.microsoft.com/kb/2264080

If your servers are set for automatic updates – you’re probably OK.  In my lab, I had them set for manual updates and I hadn’t done anything since sometime in June.  There is a very recent rollup that addresses this specific issue.  I just patched my machine (after running WU, the rollup package was in the list) and hopefully we’re done with this.

Another option is to install the SP1 Beta – the rollup is in there as well.

Happy patching!!

I wanted to build a box to dance a little with RemoteFX.  Turns out, the spare I have is running an older CPU that doesn’t have SLAT support which means ‘no bueno’. (In my case an Intel 6600 Quad Core)

Here’s more detail on what you need to run RemoteFX from a hardware perspective:

RemoteFX for RD Virtualization Host server hardware requirements

There are several hardware requirements that must be met when deploying a RemoteFX server:

• SLAT-enabled processor – The processor in the RemoteFX server must support Second-Level Address Translation (SLAT). In virtualization scenarios, hardware-based SLAT support improves performance. On Intel processors, this is called Extended Page Tables (EPT), and on AMD processors, it is called Nested Page Tables (NPT).
• GPU - At least one graphics processing unit (GPU) is required on the RemoteFX server. The GPU driver must support DirectX 9.0c and DirectX 10.0. If more than one GPU is installed in the RemoteFX server, the GPUs must be identical. The GPU must have sufficient dedicated video memory that is separate from system memory.

Also worth nothing:

For a list of GPUs that will work with RemoteFX in Windows Server 2008 R2 with SP1, see this blog post (http://go.microsoft.com/fwlink/?LinkID=197416). The list of GPUs will grow and evolve for the final release of Windows Server 2008 R2 with SP1. For a list of recommended GPU drivers, see this blog post (http://go.microsoft.com/fwlink/?LinkID=197417).

And a final important point:

In order to use Live Migration, the source and destination RemoteFX servers must have the same GPU installed in the server.

All this can be found here: http://technet.microsoft.com/en-us/library/ff817602(WS.10).aspx