Month: January 2009

There’s always questions about virtualising SQL and Exchange servers.  We know that Hyper-V performs really well and the advice is that you should be concerned about virtualising something that needs too much CPU or disk I/O.  They’d be candidates for physical deployments.  Everyone’s idea of large and heavy is relative.  You might consider your SBS server to be heavy weight.  You might consider a 4 server Exchange 2007 server deployment to be heavy.  MS thinks of that as small!

The Exchange team has done a good post on virtualising Exchange 2007 and even give some sample architectures.  They give 3 interesting solutions too, from the “small” deployment, virtual DR and an office in a can.

I just read on Hypervoria that the PowerShell library on Codeplex was updated.  Everything in Hyper-V can be managed used PowerShell.  Everything VMM does is translated into Powershell (and the scripts can be saved to the library for later use or copied/pasted).

“At present there are 80 functions in the library, some of these are worker functions which are not expected to be called directly, the others are listed below

Finding a VM
Get-VM, Choose-VM , Get-VMHost

Connecting to a VM
New-VMConnectSession

Discovering and manipulating Machine states
Get-VMState , Set-VMState , Convert-VmState,
Ping-VM , Test-VMHeartBeat, Shutdown-VM , Start-VM, Stop-VM, Suspend-VM
Get-VMKVP, Add-KVP, Remove-KVP, Get-VMJPEG

Backing up, exporting and snapshotting VMs
Export-VM , Import-VM, Get-VMSnapshot, Choose-VMSnapshot , Apply-VMSnapshot , New-VMSnapshot ,Remove-VMSnapshot, Rename-VMSnapShot, Update-VMSnapshot, Get-VMSnapshotTree, Get-VmBackupScript

Adding and removing VMs, configuring motherboard settings.
New-VM , Remove-VM , Set-VM , Get-VMCPUCount, Set-VMCPUCount, Get-VMMemory, Set-VMMemory, Set-VMSerialPort

Manipulating Disk controllers, drives and disk images
Get-VMDiskController
Add-VMSCSIController , Remove-VMSCSIcontroller
Get-VMDriveByController , Add-VMDRIVE , Remove-VMdrive
Get-VMDiskByDrive, Add-VMDISK , Set-VMDisk, Get-VMDisk
Get-VMFloppyDisk , Add-VMFloppyDisk
Add-VMNewHardDisk

Manipluating Network Interface Cards
Get-VMNic , List-VMNic , Choose-VMNIC, Add-VMNIC, Remove-VMNIC , Set-VMNICAddress , Set-VMNICConnection , Get-VMNicport ,
Get-VMnicSwitch, Choose-VMSwitch, New-VMSwitchPort, Get-VMByMACaddress, Choose-VMExternalEthernet,
New-VMExternalSwitch, New-VMInternalSwitch,New-VmPrivateSwitch

Working with VHD files
Get-VHDDefaultPath, Get-VHDInfo, New-VHD, Compact-VHD, Test-VHD,Convert-VHD,Merge-VHD,Mount-VHD, Unmount-VHD”

I still use the Hyper-V console to do some stuff because I find it suits me from time to time, despite having VMM 2008.  One of these is disk management on the cluster because you can’t remotely manage files on a GUID drive.  I just did some work on a VHD and went back to VMM to add some SCSI disks.  However, my boot VHD, an IDE disk was gone, disappeared, vamoose!

I tried a few things (checking the file location in Hyper-V MMC, refreshing the host and migrating the VM) but the fix was to make a simple alteration to the VM configuration in VMM.  I added a SCSI controller and saved the changes.  The disk reappeared.  Note: The disk was there.  Starting the machine up may have popped it back in VMM (I was in the process of deploying and customising).  I know for certain that Hyper-V knew it was there and the machine would have booted OK.

Microsoft has published a guide or instructions on how to do Live Migration (or VMotion if you’re stuck on VMware lingo) with Hyper-V as found on Windows Server 2008 R2.

You can read about the architecture here.

Your architecture options for Hyper-V aren’t just limited to a single hosted server or a single site cluster.  They won’t cover you for DR.

There’s lots of options out there.  If you have huge budget and diskless hosts, you can replicate a SAN from site A to site B across dark fibre.  Initiating the DR site is easy.  Ensure site A is offline and power up site B.  Because the hosts are diskless, they store everything about themselves (operating system, identity, configuration, services and data) on the replicate SAN.  As long as the site B hardware is identical you can power it up and carry on working as before.

That sounds perfect, eh?  But dark fibre is expensive!  You might be able to get away with 1GB copper but that’s really pushing it for SAN replication.

More affordable is host based replication.  I just saw that Double-Take has a few solutions.  Note that HP customers can get HP rebranded Double-Take software.  You can replicate VM’s on a single host to another host using Double-Take for Windows.  Or you can use GeoCluster to get much more advanced solutions.

This also reminds me that you can use things like Double-Take or Replistor to do P2V DR of data (not operating systems and applications), e.g. you can have a production physical SQL box called SQL1 and replicate the SQL databases and log files to SQL2 which is an identically configured VM in the DR site.

An interesting way to do is to continually P2V your machine using an imaging solution that can handle drivers.  Acronis have a great reputation for driver substitution in their imaging solutions.  Their True Image Echo Enterprise Server solution allows continual P2V of physical machines to virtualised DR machines.  Invocation is a restoration of your stored images, something you can regularly test in Hyper-V with private networks.  That Acronis product supports lots of operating systems.  There’s a cheaper version for Windows.

I really don’t get this whole “storage is cheap” stuff.  Storage to stick in the back of your PC is cheap but server storage is far from cheap.  The cost of a 146GB disk may be a fraction of what it was 5 years ago but we need to use a heck of a lot more to do the same stuff now as we did back then, e.g. look at the size of a Windows boot drive now.  Someone has to pay for this, e.g. an ever decreasing IT budget (IT is a non-profit generating cost centre which isn’t so popular these days) or a customer (either internal or external) has to pay for it and we all know that customers don’t want to pay for something they don’t directly use, e.g. a VMM library.  Then ask yourself, why would you fill that library LUN with files that are 75% empty?

If you’ve been using VMware ESX then you know how efficient it is when it comes to storing template machines.  Unfortunately, VMM 2008 isn’t that intelligent.  I tried to have this conversation with one of the folks at TechEd but because of MS’s tendency to rename every industry accepted term he hadn’t a clue at what I was trying to say.

So here it is in VMware (sort of) lingo and then I’ll translate.  In Virtual Center, you can sysprep a machine (either via the GUI which I do not recommend because sysprep is both OS and service pack specific which VC cannot handle) or manually (which I do recommend).  When I covert that VM into a template it’s removed from the production VMFS LUN and placed into the library.  Part of this (if I remember correctly – it’s been a while) was that the disk is stored as a dynamically expanding disk instead of a fixed size one.  Why store 40GB of mostly empty virtual disk when you can store 8GB?  Imagine having lots of templates that you need to manage.  You could have a library that’s TB upon TB of wasted space.  When you deploy the template, the virtual disk is converted back to fixed size on the VMFS LUN.

So I asked about taking a template machine in VMM and doing the same.  Straight away said person got confused because MS decided template would refer to the simple file that describes the VM configuration and nothing else.  The conversation beyond that point could go nowhere.  So here’s what would be great.  Take a VM and sysprep it down.  The library storage process would take the VHD on the Hyper-V host and if it’s fixed it would convert (without an unnecessary copy taking up space) over the wire to a dynamically expanding VHD in the library.  A deployment of that VHD would ask if you want a fixed or dynamic disk.  If it’s fixed then the VHD would be converted, again over the wire without a wasteful copy, and a new fixed size VHD would appear in the destination LUN on the Hyper-V host.

Here’s What I Do

That’s not there now so I’m doing something slightly more manual.  I build a template (yeah I said “template” cos that’s what it is; Ask any OS deployment person) using a dynamically expanding VHD on my development host.  I then copy it into my library which is a compressed folder.  Sure it’s a slight bit slower but it’s worth the saved space.  For a VM deployment I deploy a VMM template (ICK – the machine configuration kind) to the Hyper-V host’s LUN for that VM.  I log into the host and then fire up the Hyper-V console.  Using the disk editor I browse to the VMM library and convert that disk to a fixed type whose location is the VM’s LUN. 

That manual process does everything that I’d like MS be able to do with VMM.  It might take me about 1 minute more to deploy a VM than it would otherwise but I’m saving a tonne of space on expensive disk.

This is the only thing that really bugs me with VMM.  But it really annoys me.  I’m guessing the mix between the restrictive Hyper-V security model and the inability to access GUID drives over the network have caused this.  The new cluster file system in Hyper-V R2 will give them an opportunity to sort this out.  Hopefully MS will sort it out. 

It’s been a long time since I wrote anything outside of work.  Work projects really ate up a lot of time and pretty much exhausted me over the last 2 years.  I’ve just spent the last 2 hours planning and writing again for the first time in ages.  It might surprise some (not) to find out that the subject in question is Hyper-V.  There’s no schedule and no contracts so I’ll post the document when I post it.  I don’t think this one will take me too long; I hope it won’t because the old speaking schedule is stacking up.

EDIT (23:50 31st Jan 2009):

I’ve gotten 18 pages written so far over bits of 3 days.  Good progress so far.  Probably 50% done with this document.