The cmdlets can be downloaded from VMware’s website as a PowerShell Snapin and require PowerCLI 4.0 U1.

I look forward to seeing a few scripts pop out in the next few days that make use of these.

• Matt Deacon – Microsoft
• Simon Wardley – Canonical
• Rod Johnston – VMware
• Chris Richardson – Thoughtworks

I’m not particularly familiar with any of them but it could be interesting having VMware and Microsoft squaring up against each other to talk about Clouds.

As an added bonus it’s just round the corner from where I’m working. I think I’ll go and register now. For more details (including the registration process) see CloudCamp’s site.

Date:

Thursday, March 11, 2010 from 6:30 PM – 11:00 PM (GMT)

Location:

Queen Elizabeth II Conference Centre
Broad Sanctuary
Westminster
SW1P 3EE London
United Kingdom

Actually I registered a few hours ago now but something has been bugging me. The big reveal is due on March 22nd. There are still 9 days of February to go. Doing the maths, 22 + 9 does not equal 23 (see the image below). It’s a lot closer to 32. Perhaps Veeam have a small code problem? I’d hope not and I think not.

Having had a look at the page source,

it looks like the countdown is configured to run until March 15th. I’m guessing that there will be more revealed or some change to their site around that time. I wait in eager anticipation.

Update (20/02/2010): Found a tweet from @veeam this morning:

RT: @veeam: RT @mpoore: New Post: Veeam Surebackup http://bit.ly/bAVtum <- Good detective work…the countdown will be changed for March 22 (OOPS)

For a quick answer, run this VBScript:

set objSysInfo = CreateObject("ADSystemInfo")
set objUser = GetObject("LDAP://" & objSysInfo.UserName)
wscript.echo "DN: " & objUser.distinguishedName

Ah, found it. Of course, things would be much quicker with PowerShell but 4 hours ago I was the new guy and PowerShell doesn’t really exists that side of the firewall yet

Every time in that session that “Connect-VIServer” is used I’ll get it because my normal domain account doesn’t have any privileges in vCenter. The point of this post isn’t to teach anyone to suck eggs but maybe it’ll help me remember in future to make one tiny change after installing PowerCLI.

Right click on the PowerCLI shortcut and open the shortcut’s properties.

Click the “Advanced” button.

Tick the box “Run with different credentials”. OK everything.

The next time the the shortcut is used the option to specify different credentials is shown:

Entering a different account here saves having to do it each time “Connect-VIServer” is used. There are other things that can be done of course. Maybe I’ll save those for another day.

• Localisation of the View client in a number of languages including French and German.
• Support for vCenter 2.5 Update 6 and ESX 3.5 Update 5
• PCoIP now supports Virtual Printing, a couple of Single Sign-On Providers and International keyboards.

There are a few bug fixes included also. The best thing to do is to head over to the release notes and check them out. Then it’s time to download and update .

After consulting the PowerCLI cmdlets reference I found out exactly how many there were.

(get-VIPrivilege).count

Only 249. Multiply that by 14 default roles and you get 3486 permissions to document. It’s not quite that bad really. You only need to document permissions that have been assigned so that cuts the number down considerably. Still, there must be an easier way.

I’ve all but hit you over the head with the solution. Let’s take the default “Admin” role as an example. The following shows exactly which permissions it has:

get-VIPrivilege -role Admin

The output isn’t pretty but it gets the job done. Suppose you want the privileges for all roles. To borrow Apple’s catch-phrase, there’s a cmdlet for that.

get-VIRole

This is all well and good but we need to combine these cmdlets, their output and present it in some sort of meaningful way. PowerShell has a number of builtin cmdlets for formatting and / or exporting data. What I wanted was something that didn’t require me to do any extra formatting afterwards and that I can use again. Given the nature of the data we’re trying to extract, one of the best ways of presenting it is in an Excel spreadsheet. For this we could use the export-csv cmdlet but to avoid any extra formatting I decided to experiment with Excel interaction. That is the part that took most of the time and has made the script below such an ungainly beast. Thanks though to the magic of Google and inspiration from one of Alan Renouf’s scripts, I cobbled together the following:

####################################################################
# List-Privileges.ps1                                              #
#                                                                  #
# Author: Michael Poore (www.wekabyte.co.uk)                       #
# Version: 0.1                                                     #
# Date: 12/02/2010                                                 #
#                                                                  #
# Change History:                                                  #
# - 0.1 - First working version                                    #
#                                                                  #
####################################################################

$vcserver = "myvcserver"
$startrow = 3
$date = get-date -format F

# Connect to VC Server
Connect-VIServer $vcserver

# Get a list of all privileges from the VC Server
$privs = @()
foreach ($priv in Get-VIPrivilege | sort Id)
{
 $objecta = ""  | select-Object ID,Description
 $objecta.ID = $priv.Id
 $objecta.Description = $priv.Description
 $privs += $objecta
}

# Get a list of all roles from the VC Server and determine which privileges they hold
$roles = @()
foreach ($role in Get-VIRole)
{
 $objectb = "" | select-Object Name,System,Description,Privileges
 $objectb.Name = $role.Name
 $objectb.System = $role.IsSystem
 $objectb.Description = $role.Description
 $myprivs = @()
 $roleprivs = $role.PrivilegeList | Sort
 $roleprivs
 foreach ($priv in $privs)
 {
 $myprivs += $roleprivs -contains $priv.ID
 }
 $objectb.Privileges = $myprivs
 $roles += $objectb
}

# Create new Excel object
$Excel = New-Object -Com Excel.Application
$Excel.visible = $True
$Excel = $Excel.Workbooks.Add(1)
$Sheet = $Excel.WorkSheets.Item(1)

# Write Worksheet title
$Sheet.Cells.Item(1,1) = "Roles and Privileges Report for $vcserver - $date"
$Sheet.Cells.Item(1,1).font.bold = $true
$Sheet.Cells.Item(1,1).font.underline = $true
$Sheet.Cells.Item(1,1).font.size = 18

# Write worksheet column headers
$row = $startrow
$Sheet.Cells.Item($row,3) = "ROLE:"
$Sheet.Cells.Item($row,3).font.bold = $true
$Sheet.Cells.Item($row,3).HorizontalAlignment = 4
$Sheet.Cells.Item($row,3).Borders.Item(10).LineStyle = 1
$Sheet.Cells.Item($row,3).Borders.Item(10).Weight = 4
$row++
$Sheet.Cells.Item($row,3) = "DESCRIPTION:"
$Sheet.Cells.Item($row,3).font.bold = $true
$Sheet.Cells.Item($row,3).HorizontalAlignment = 4
$Sheet.Cells.Item($row,3).Borders.Item(10).LineStyle = 1
$Sheet.Cells.Item($row,3).Borders.Item(10).Weight = 4
$row++
$Sheet.Cells.Item($row,3) = "SYSTEM:"
$Sheet.Cells.Item($row,3).font.bold = $true
$Sheet.Cells.Item($row,3).HorizontalAlignment = 4
$Sheet.Cells.Item($row,3).Borders.Item(10).LineStyle = 1
$Sheet.Cells.Item($row,3).Borders.Item(10).Weight = 4
$Sheet.Rows.Item($row).Borders.Item(9).LineStyle = 1
$Sheet.Rows.Item($row).Borders.Item(9).Weight = 4
$row++
$sheet.columns.item(1).columnwidth = 5
$sheet.columns.item(2).columnwidth = 5
$sheet.columns.item(3).columnwidth = 30
$sheet.columns.item(4).columnwidth = 2

foreach ($priv in $privs)
{
 $level = [regex]::matches($priv.ID,"\.").count
 switch ($level)
 {
 0 {$col = 1}
 1 {$col = 2}
 default {$col = 3}
 }
 $Sheet.Cells.Item($row,$col) = $priv.Description
 $Sheet.Cells.Item($row,3).Borders.Item(10).LineStyle = 1
 $Sheet.Cells.Item($row,3).Borders.Item(10).Weight = 4
 #$Sheet.Cells.Item($row,3).WrapText = $true
 $Sheet.Cells.Item($row,4) = " "
 $row++
}

$col = 5
foreach ($role in $roles)
{
 $row = $startrow
 $Sheet.Cells.Item($row,$col).Orientation = 90
 $Sheet.Cells.Item($row++,$col) = $role.Name
 $Sheet.Cells.Item($row,$col).HorizontalAlignment = 3
 $Sheet.Cells.Item($row++,$col) = $role.System
 $Sheet.Cells.Item($row,$col).HorizontalAlignment = 5
 $Sheet.Cells.Item($row++,$col) = $role.Description
 foreach ($priv in $role.Privileges)
 {
 if ($priv)
 {
 $Sheet.Cells.Item($row,$col).HorizontalAlignment = 3
 $Sheet.Cells.Item($row,$col) = "Yes"
 }
 $row++
 }
 $sheet.columns.item($col).columnwidth = 6
 $col++
}
Clear

It’s a little slow to run but fairly easy to read the output. These are the default roles in vCenter 4.0.

So that’s part 1. The next step is already underway – tidying the script and adding to it to try and give some indication of which AD users and groups map to which roles.

The solution is a little easier than that though. The key thing is using the right file. The VMware-viclient.exe file that can be downloaded from the vCenter server or an ESX / ESXi host is a self-extracting archive and doesn’t pay any attention to the normal parameters and switches that you can try to add to get a silent installation performed.

The right file to use is from the vCenter installation media (zip file or iso). In the “vpx” folder is a VMware-viclient.exe file that is 137Mb in size (the archive file mentioned above is about 114Mb). If you place the file on the root of your C:\ drive then a silent installation (an unattended one at least) can be performed using the following command line:

start /wait C:\VMware-viclient.exe /q /s /w /L1033 /v" /qr"

If you’d like a log file created, use this instead to write a logfile to the user’s temp directory:

start /wait C:\VMware-viclient.exe /q /s /w /L1033 /v" /qr /L*v \"%TEMP%\vmvcc.log\""

After some poking around I found that the local Administrators group did not not have Full Control of the registry key HKLM\Software\Microsoft\Windows NT\CurrentVersion\Svchost to which it was trying to write.

Just thought I’d share this in case anyone else encountered it.

VMware explain their support lifecycle policies here and basically the clock started ticking for VI3 support the day that vSphere went GA. VI3 moves into extended support on 21st May 2010. That’s only 4 months away.

So what does Extended Support mean? This is VMware’s definition:

New hardware platforms are no longer supported, new guest OS updates may or may not be applied, and bug fixes are limited to critical issues. Critical bugs are deviations from specified product functionality that cause data corruption, data loss, system crash, or significant customer application down time and there is no work-around that can be implemented.

I’m going to stick my neck out a bit and suggest that this may be the final update to vCenter 2.5 before it reaches the end of General Support in May.

Read the full article here: Breaking down the vStorage Virtual Machine File System (VMFS)

Ooops. I had a good rummage in the hostd.log file on the host that attempted to power on the VM and found the following messages:

[2010-01-27 12:33:09.101 'BaseLibs' 133225392 info] DISKLIB-VMFS : "/vmfs/volumes/4a081291-4fb12f12-bef0-001e0bcdc996/myvm/mydisk_1-000001-delta.vmdk" : open successful (21) size = 16106127360, hd = 0. Type 8
[2010-01-27 12:33:09.103 'BaseLibs' 133225392 info] DISKLIB-VMFS : "/vmfs/volumes/4a081291-4fb12f12-bef0-001e0bcdc996/myvm/mydisk_1-000001-delta.vmdk" : closed.
[2010-01-27 12:33:09.151 'BaseLibs' 133225392 info] SNAPSHOT: Unable to find all files for '/vmfs/volumes/4992b455-063c9aec-5e36-001e0bcdc996/mytemplate/mydisk_1.vmdk'
[2010-01-27 12:33:56.219 'vm:/vmfs/volumes/4a081291-4fb12f12-bef0-001e0bcdc996/myvm/myvm.vmx' 20868016 info] Question info: VMware ESX Server cannot find the virtual disk "/vmfs/volumes/4992b455-063c9aec-5e36-001e0bcdc996/mytemplate/mydisk_1.vmdk". Please verify the path is valid and try again.
Cannot open the disk '/vmfs/volumes/4a081291-4fb12f12-bef0-001e0bcdc996/myvm/mydisk_1-000001.vmdk' or one of the snapshot disks it depends on.
[2010-01-27 12:33:56.240 'ha-eventmgr' 20868016 info] Event 81 : Message on myvm on myhost.local in ha-datacenter: VMware ESX Server cannot find the virtual disk "/vmfs/volumes/4992b455-063c9aec-5e36-001e0bcdc996/mytemplate/mydisk_1.vmdk". Please verify the path is valid and try again.
Cannot open the disk '/vmfs/volumes/4a081291-4fb12f12-bef0-001e0bcdc996/myvm/mydisk_1-000001.vmdk' or one of the snapshot disks it depends on.

(I’ve sanitised this log file snippet so the names aren’t accurate but they are consistent with the issue that I discovered.)

Firstly the logfile shows a delta file. That means that the VM is running from a snapshot. This didn’t show up beforehand and the Snapshot Manager did not show it. Most likely VCB (or the backup software using it) didn’t clean up after itself. Browsing the datastore where the VM resides showed that the snapshot was nearly two weeks old.

Secondly, you can see the issue in the third line onwards. It looks like the base disk file has gone missing. However reading more closely it looks like the base disk is on a different datastore and actually part of a different VM! For some reason, when this VM was deployed from a template it retained one of the template’s disks as its own. Looking into that datastore I could see the mydisk_1-flat.vmdk file but there was no mydisk_1.vmdk file. (Just to explain, the former is the actual disk file. 15Gb in size and containing the VM’s data. The latter file is a small text file and contains configuration data. I’ll call it the disk descriptor file.) So, it was a missing disk descriptor file that was the issue. I did a quick google and didn’t find anything immediately helpful so I ran through the following steps:

1. Copied the mydisk_1-flat.vmdk file from the template VM’s datastore to the broken VM’s datastore.
2. Knowing that the disk was supposed to be 15Gb in size, I created a quick VM with a single 15Gb disk and copied the disk descriptor file to the broken VM’s datastore.
3. Next I made a note of the parentCID from the mydisk_1-000001.vmdk disk descriptor file. This value (from the snapshot delta’s disk descriptor file) is the ID of the parent disk.

# Disk DescriptorFile
version=1
CID=de54d5dd
parentCID=1bb73626
createType="vmfsSparse"
parentFileNameHint="/vmfs/volumes/4992b455-063c9aec-5e36-001e0bcdc996/mytemplate/mydisk_1.vmdk"
# Extent description
RW 31457280 VMFSSPARSE "mydisk_1-000001-delta.vmdk"

# The Disk Data Base
#DDB

ddb.toolsVersion = "7302"

4. I also modified the file above to correct the parentFileNameHint value so that it referred to the local datastore and became:

parentFileNameHint="mydisk_1.vmdk"

5. I modified the newly created 15Gb disk descriptor file with the CID matching the parent value from step 3. And made sure that the Extent description was correct.

# Disk DescriptorFile
version=1
CID=1bb73626
parentCID=ffffffff
createType="vmfs"

# Extent description
RW 31457280 VMFS "mydisk_1-flat.vmdk"

# The Disk Data Base
#DDB

ddb.virtualHWVersion = "4"
ddb.uuid = "60 00 C2 91 07 97 77 cb-87 9e 5d 9f 95 95 2c 46"
ddb.geometry.cylinders = "1958"
ddb.geometry.heads = "255"
ddb.geometry.sectors = "63"
ddb.adapterType = "lsilogic"

6. I saved the file as mydisk_1.vmdk

The VM then powered on successfully. I checked the disks after successful boot up and they’re there.

Now all that remains is to sort out the snapshot. It still doesn’t register in snapshot manager.

This has been a bit of a hack but it worked. And before anyone comments, I just modified my google search terms and found the answer in a VMware KB – first hit! Recreating a missing virtual disk (VMDK) header/descriptor file

CloudCamp

Format

The format of the evening was quite refreshing. Six lightning talks by various speakers to be followed by a question panel and focused break-out sessions. (See the full agenda here). The lightning talks in particular were slightly new to me but very effective. Having just five minutes to talk about a subject without being too vendor specific can’t be easy. In a minimalist way it forces the speaker to focus on the important things. Of the six speakers / subjects I found that Kate Craig-Wood’s five minutes on the “UK G-Cloud” got me thinking the most as I have done a lot of work with central and local government bodies.

Unpanel

The question panel (“unpanel”) featured six volunteers giving their thought / answers about various questions posed by the other audience members. Again the idea is not to be too vendor specific and a red card system was in place to send off any of the panel who strayed too far in one direction or another. There were some interesting questions asked and some very good answers given. Plenty for me to think about on my train home.

Breakout Sessions

The topics for the sessions were nominated by anyone with an idea for one. “The role of the government in the Cloud” was booed as a suggestion and “The role of Microsoft in the Cloud” was met with stony silence. That the session on “Interoperability and Standards” was chaired by a man from Microsoft was met with several undisguised chuckles.
I had meant to attend to join in with one of the discussions but after a drinks / beer break I was chatting about all sorts with bloggers Simon x 3 (Gallagher, Long and Seagrave of vinf.net, simonlong.co.uk and techhead.co.uk respectively) and Stuart (vinternals.com).

Thoughts on Clouds

I’ve had numerous thoughts about Cloud Computing in recent months and more after last night. At some point I might write some of it down and share it.
As for CloudCamp, I enjoyed it and I definitely got something from it. I might not attend every one in the future but I’d certainly recommend it for anyone with an interest or potential interest in Cloud Computing and the future of virtualisation.

The proposed agenda (shamelessly pinched from the announcement page on VMware Communities) is:

1100 – 1200 (Optional) PowerCLI / Powershell workshop – Alan Renouf. Please bring your own curly brackets.

12:30 – 13:00 Arrive & Refreshments
13:00 – 13:20 Welcome & News – Alaric Davies

13:20 – 14:00 Sponsor Presentation – Chris Hammans, Pano Logic

Real world vSphere deployment experiences – Stuart Thompson
(Mostly) Zero downtime DC migration for Dummies – Jonathan Medd

15:00 – 15:20 Refreshment break

ESX home lab update, virtualizing Terminal Server workloads – Simon Gallagher
Thin provisioning and capacity planning in a virtual world – Chris Evans, ‘The Storage Architect’

16:45 – 17:00 Close
17:00 – Pub

I plan to be there.

Let’s start with what I want to accomplish. I’m working with an ESX 3.5 web hosting environment and there’s a new project in the pipelines. A brace of new servers are required and they’ll be on a new VLAN. The VLAN has been created and configured on the various switches that the ESX hosts connect to but now of course a corresponding PortGroup is required. (Actually two are needed – complex project.) Now we’re not talking about a huge number of hosts here. It would probably only take 10 minutes to do it by hand using the VI client. It’ll take me longer to write this post! However, it is something that happens relatively often in this environment so it’s worth taking the time to write a script.

Let’s say, for the sake of simplicity, that I have 5 or 6 hosts with near identical networking configurations. The same number of NICs in each host, the same NICs attached to the same vSwitches in each host.

The command that we’ll need to use to create the PortGroup will be New-VirtualPortGroup. We’ll need to give it a name for the PortGroup and a VLAN ID too. Crucially it will need to know which vSwitch we want the new PortGroup to be placed on. Many environments will have more that one vSwitch and PowerCLI won’t be able to guess for you. The VirtualSwitch parameter of the New-VirtualPortGroup cmdlet needs to be an object so we have to use the Get-VirtualSwitch cmdlet with sufficient detail to identify the vSwitch that we want uniquely. Using the vSwitch name and the VMHost gives a unique combination.

Finally, to add the PortGroup to all hosts in a cluster we have to call the New-VirtualPortGroup cmdlet once per VMHost. This can be done using the PowerShell pipeline.

So, now that I’ve confused you with a few parapgraphs of waffle, adding a new PortGroup is actually pretty simple. In fact it could be a one liner once you are connected to vCenter. Something like this:

Get-Cluster -Name $cluster | Get-VMHost | foreach { New-VirtualPortGroup -VirtualSwitch ( Get-VirtualSwitch -Name $vswitch -VMHost $_ ) -Name $PG -VLanId $vlan }

Of course you’d need to insert your own values for $cluster, $vswitch, $PG and $vlan as follows:

$cluster = The name of the cluster in vCenter that your hosts are in

$vswitch = The name of the vSwitch that you want to add the PortGroup to

$PG = The name of the PortGroup that you want to add

$vlan = The VLAN ID that the PortGroup should use (0 can be used to indictae no VLAN ID)

It should guarantee that if you make any spelling mistakes / typos that they are at least consistent across your infrastructure You could tack -whatif onto the end of the New-VirtualPortGroup command to test it before making any changes. As always, be careful when you make scripted changes. If possible test them in a development environment first.

To take this further, you could convert the whole thing into a script that takes input from the command prompt or as parameters.

I first heard about the book in the run up to VMworld 2009 back in September. I would very much have like to go but could not. I gather that some preview copies of the book were given away there. Luckily I know and live quite close to one of the authors and managed to get my hands on a copy. I can’t recall what my expectations were but the book that I started reading that night was easier to read and more compelling than anything that I’d picked up for a while.

The structure of the book is fairly loose – there are no chapters. It is broken down into different sections covering vCenter, Networking, Storage etc and in turn each of those is broken down into sub topics. These take the form of FAQ type questions that are then answered or are descriptive sections of text. Throughout the book are helpful scripts or commands that can be used, the majority being written in PowerShell.

The quality of the content is very high and should appeal to seasoned vExperts and relative beginners alike. While reading it I certainly found areas where my knowledge was rusty or non-existent and many other areas where the book will serve as a handy reference when I can’t remember something. And given the size of the book (~240 pages and pocket sized) it’s very easy to carry around.

There is talk of further titles coming from the same authors and if this volume is indicative of the quality that we can expect then I am looking forward to seeing more. I also hope that it inspires other people to consider sharing their knowledge like this.

The environment that I’ve been working in is a shared web hosting platform built on ESX 3.5 and vCenter 2.5. Recently the focus of effort has been on data centre migrations, service migrations and office moves. Business as usual (BAU) activities have taken a back seat because of resource constraints. That includes upgrading the virtual infrastructure to the latest version. Currently vCenter is at Update 4 (build 147633) and needs to be brought up to date following the release of Update 5 a few weeks ago. The ESX hosts are slightly further downlevel however. Two thirds of them are on Update 3 (build 123630) with a couple of extra patches applied to bring them up to build 143128. The other third is on Update 2 (build 110268).

Windows Server 2008 R2 has recently started to be used for some of the VMs and this is not officially supported until Update 5. Some of these VMs have had minor issues, things like random slowness and occasional failures to remove snapshots following their scheduled backups. While looking into these issues I discovered quite a variety of VMTools versions being used.

Ideally all ESX hosts in a cluster should be on the same update level. The version of VMTools that a VM uses should ideally match the lowest build version of ESX that is being used in a cluster as the VM could potentially run on any host in the cluster. On one VM vCenter said that the VMTools version was “OK” but I wanted to run a manual “Install / Upgrade” on the VMTools anyway just to run a repair and make sure that the drivers were installed correctly. When I tried to do this, the installer popped back an error stating that a newer version was already installed.

That got me thinking, which versions are installed on all the VMs? Now I could have used the excellent RVTools or something like that or I could craft my own brief PowerCLI script. I couldn’t resist the latter…

$vmhosts = get-vmhost | sort Name

$myCol = @()
foreach ($vmhost in $vmhosts)
{
   $vmhostdetail = $vmhost | get-view
   $versionmajor = $vmhostdetail.config.product.version
   $versionbuild = $vmhostdetail.config.product.build
   $vms = $vmhost | get-vm | sort Name
   foreach ($vm in $vms)
   {
      $myObj = "" | Select Host,Version,Build,Guest,ToolsVersion
      $vmdetail = $vm | get-view
      $vmtoolsversion = $vmdetail.config.tools.toolsversion
      $myObj.Host = $vmhost.Name
      $myObj.Version = $versionmajor
      $myObj.Build = $versionbuild
      $myObj.Guest = $vm.Name
      $myObj.ToolsVersion = $vmtoolsversion
      $myCol += $myObj
   }
}

$myCol | Export-Csv -NoTypeInformation "C:\VMToolsVersions.csv"
ii "C:\VMToolsVersions.csv"

Running this script produces a CSV file with a line for each VM show the host, host version and build number, guest name and VMtools version for each VM. I’ve simplified the output that I got to remove the host names and guest names and also I’ve removed duplicate combinations of version numbers. That just leaves the following:

This shows the different combinations of ESX hosts versions and VMTools versions being used. What’s very interesting are the VMs running VMTools version 164009. That is the build number for vSphere 4.0.0!

I know how this has come about. Rather than using templates, the preference here is to use a Windows ISO file that has been customised to be a practically unattended install that also includes a number of automatically installed applications. One of these is the VMTools installation. As there is a vSphere infrastructure here also the ISO has been modified to include the latest tools version so that one ISO fits all. But does it? The VMs are running butis the version of VMTools causing problems? That’s the bit that I don’t know the answer to yet. If anyone does know, please feel free to comment.

For now the way forward is going to be creating multiple ISO files or removing the VMTools installation from the ISO altogether. That way new builds will get their VMTools installed from the ESX server that they’re running on. An upgrade to Update 5 is also called for and with next week looking to be a bit quieter I think that it’ll be planned then to be executed early in the new year.

Initially I was asked about what was causing a SQL VM to respond slowly and use 100% CPU. I had a look in vCenter and while it looked slightly busy it didn’t seem over worked. As the graph to the right shows, it was using only about half of the available 3Ghz CPU it had access to. Perhaps I should explain further at this point that my client’s practice when it comes to VMs is to provision them with a single vCPU and add more if they are required. It seems that if this was normal load for the VM that 1 vCPU should be enough.

Looking back through the VM’s performance history I could see nothing particularly wrong either. Occasional CPU spikes in the past possibly indicating reboots or overnight processing. Oddly though there were random plateaus of activity for several hours at a time. Mostly overnight the VM would idle along using practically no CPU resource but during the day there were long periods where it looked a lot like the activity above. Time to look at the guest OS.

The picture inside Windows is slightly different. Opening up task manager shows frequent bursts of 100% CPU usage (see below). Actually you could call them regular. And, more worringly, it transpires that the server is not yet in production – it’s still being configured.

The offending process is services.exe so it’s not immediately obvious what the issue is. Purely by coincidence I asked the DBA if she could log off for a while so that I could look into what was going on. When she did, the strangest thing happenned:

See how the CPU usage dropped back down to idle and stayed there. That begged the question “What were you running?”.

It turns out that the culprit was none other than SQL Management Studio. When opened and connected it polls the server’s status every 10 seconds. Strangely though, instead of polling just the SQL services it polls all services on the server (this can be seen using Process Monitor) which seems a bit excessive to me. Due to the way that hypervisors share resources, what would be a small blip on a physical host is magnified within the VM somewhat. Microsoft have acknowledged that this happens but to my knowledge haven’t done much about it. There is a registry key that can be modified to adjust Management Studio’s behaviour. For SQL 2005 SP1 onwards (it’s not available before that) it is:

HKLM\Software\Microsoft\Microsoft SQL Server\90\Tools\Shell\PollingInterval

Setting it to 600 will reduce the frequency of polls to once a minute. Alternatively, just don’t leave SQL Management Studio open longer than you have to and wait for Microsoft to fix it.

After downloading it locally, use the vSphere client to mount the ISO file on your VM. From within the command prompt on the guest OS, change drive to your CD / DVD drive (D: in my screenshot below) and just run the”Start_Coreconfig.wsf” file.

This opens a GUI with a few options for configuring Server Core.

I’m not going to go into too much detail about what can be done, try it yourself. But just as an example, compare the following window with the steps to configure networking using the command line.

Fantastic tool. I’ll be adding it to my toolkit for sure.