I made a blog post some time ago, where i created a powerCLI function for creating a VM affinity rule that would keep my VM´s in a vApp together in the cluster for better VM-VM performance. I then checked for MoRef to get the VM correlated between vCloud Director and vCenter.

The answer to my problem was much closer than I thought, the vCloud director creates a folder in the “VM´s and Template view” for each vApp and guess what, all VM´s are in there.. so I do not need to ask the vCloud Director anymore (well if I want the Cloud vApp name I do need to ask it..) but ironically I had not looked there and found out that the folders actually where created for each deployed vCloud vApp :-/

So the code is a bit simpler and to get all VM´s from a folder I just type this, the asterisk is because the folder also gets a unique id

Get-Folder | where {$_.Name -like "ProdvCPod*"} | Get-VM

And then I just need to put those VM´s into an array and create a DRS rule, and here is the function that does this

<#
.Synopsis
   Add an affinity rule for a vCloud Director vAPP
.DESCRIPTION
   This function takes a vApp name (string) as parameter and creates an affinity rule for them to keep them together
.EXAMPLE
   Add-vAppAffinity -CIVApp ProdCust01
.NOTES
Author: Niklas Akerlund
Date: 2012-06-21
#>
function Add-vAppAffinity
{
    Param
    (
        # Parameter for the vAPP
        [Parameter(Mandatory=$true,
                   ValueFromPipelineByPropertyName=$true,
                   Position=0)]
        $CIVApp,

        # If the rule should apply on a different cluster
        $Cluster = "Cluster01"
    	)

	if ($CIVApp -ne " " -or $CIVApp -ne $null -or $CIVApp.length -ne 0 ){
		$Cluster = Get-Cluster $Cluster
		$VAppname = $CIVApp + "*"
		$VMs =  Get-Folder | where {$_.Name -like $VAppname} | Get-VM	
		
		if (!(Get-DrsRule -Cluster $Cluster -Name $CIVApp -ErrorAction SilentlyContinue)){ 
			New-DrsRule -Name $CIVApp -Cluster $Cluster -KeepTogether $true -VM $VMs
		}else{
			Remove-DrsRule -Rule (Get-DrsRule -Cluster $Cluster -Name $CIVApp) -Confirm:$false
			New-DrsRule -Name $CIVApp -Cluster $Cluster -KeepTogether $true -VM $VMs
		}
	}
}

To run it:

and it looks something like this if I check the cluster settings

preferably you have set your DRS cluster to fully automated or not so much will happen 🙂

This week I have been teaching my first vSphere 5 ICM, I have been trying to show the students the wonderful world of PowerCLI and all good you can do with it.

When we got to the virtual networking I tested if I could add a switch with any number of ports

get-vmhost esxi01* | New-VirtualSwitch -Name TestNumPorts -NumPorts 12 -Nic vmnic0

When I try to add 12 ports with the -NumPorts parameter, powerCLI tells me that my switch has been created with 16 ports, but look here in my vSphere Client screen dump

hmm smells like a bug? or ?

Well now I wanted to test if I could add more than 4 VM´s and connect them and then start them, first I have to create a Portgroup that i can connect and then a nice for loop to create 10 VM´s

get-vmhost esxi01* | Get-VirtualSwitch -Name TestNumPorts | New-VirtualPortGroup -Name Test

for ($i=1;$i -le 10;$i++){New-VM -VMHost (Get-VMHost esxi01*) -Name ("VM01" + $i) -Datastore (Get-Datastore delad02) -NetworkName Test}

The last  oneliner creates a standard VM with 256 MB ram and a 4 GB vmdk disk, and then I want to start them to actually connect them to the network I created to see if anyone of them does not get connected.

So to get them connected I start them

And check the last command where we can see that my number of available ports has not decremented?

Here you can see in the vSphere client that all my new VM´s are connected to the Test portgroup

In this screenshot you can see why my vSwitch got 16 ports (as 8 are allocated by the vmkernel) and the options are 8/24/56/120/248/504/1016 and so on, but still it does not explan why the powercli does not show that i have used 10 of the ports, or?

To see maybe what causing this I changed the number of ports to one of the “allowed” and when I changed the number to 16 ports, directly my free ports changed to 4, but I have 10 machines + 8 reserved?

The following powerCLI command shows the VM´s and if they are connected if you do not believe the gui 😉

Get-VM VM01* | Get-NetworkAdapter | select parent, connectionstate

The conclusion is that I have no answer if this is as it should be or if it is a bug, maybe someone out there could clarify 🙂

Now time to prepare for tomorrow and the modules in the course 🙂

Today I had an reason for running PowerCLI again, the case was to get an quick report on the datastores at a customer, I have made a post about the one-liner that get the number of running VM´s on a datastore. As i described in that post, if your SAN does not support VAAI then you do not want to many VM´s on each datastore because of the SCSI-locking that can occur. This is just an extension of that because we also wanted to check how overprovisioned the datastores was (when using thin provisioning there is a risk that you will fill your datastores as the VM´s fill their vmdk disk´s)

The PowerCLI code looks like this

Get-Datastore | Select Name,@{N="TotalSpaceGB";E={[Math]::Round(($_.ExtensionData.Summary.Capacity)/1GB,0)}},@{N="UsedSpaceGB";E={[Math]::Round(($_.ExtensionData.Summary.Capacity - $_.ExtensionData.Summary.FreeSpace)/1GB,0)}}, @{N="ProvisionedSpaceGB";E={[Math]::Round(($_.ExtensionData.Summary.Capacity - $_.ExtensionData.Summary.FreeSpace + $_.ExtensionData.Summary.Uncommitted)/1GB,0)}},@{N="NumVM";E={@($_ | Get-VM | where {$_.PowerState -eq "PoweredOn"}).Count}} | Sort Name | Export-Csv -Path C:\temp\vmds-datastore.csv -NoTypeInformation -UseCulture

And this is when i have run it on my test system, the difference here is that i removed the Export-CSV to get the output in the console

And here is a simple excel report

Today I wanted to update my lab environment hosts to U1 of vSphere 5, of course with PowerCLI, there is no other way or is it?

This reference indicates there is not so many cmdlets but is it enough? Yes it is!

As the PowerCLI installation does not include the VUM cmdlets i had to download and install them

I have already upgraded my vCenter and VUM to U1, this as it is always a best practice to first update the vCenter and then the hosts

the following script do the whole process of both making a Patch baseline and attaching it to the cluster and then Scan/Remidate (now i only have two hosts but this is quite powerfull if i have lots of more). The script allows some parameters, for example if you already have a baseline you can use that and attach/remediate your cluster, or you want another patch in the baseline.

<#
 .SYNOPSIS
 Upgrade ESXi Hosts to an Update Version 

 .DESCRIPTION
 Create an baseline and upgrade a Cluster to an vSphere 5 update with Update Manager, If you already have a baseline just use the BaseName parameter

 .Author
 Niklas Akerlund /RTS 2012-03-23 (original script from Jonathan Medd)

 .PARAMETER  Cluster
 The name of the host to upgrade.

 .PARAMETER  PatchName
 Enter the name of the Patch/Update
 
 .PARAMETER Name
 Enter the name of the vCenter server to connect to

 .EXAMPLE
 PS C:\> Update-HostsU1.ps1 -Cluster VMWCL
#>
#requires -pssnapin VMware.VimAutomation.Core
#requires -pssnapin VMware.VumAutomation

[CmdletBinding()]
param(
 [Parameter(Position=0, Mandatory=$true, HelpMessage="Enter the name of Cluster")]
 $Cluster,
 [Parameter(HelpMessage="Enter the name of the patch")]
 [String]
 $PatchName = "VMware ESXi 5.0 Complete Update 1",
 [Parameter(HelpMessage="Enter the baseline name")]
 [String]
 $BaseName = "vSphere ESXi 5 U1"
 )

process{

	# Create a baseline to use
	if(!(Get-Baseline -Name $BaseName -EA 0)){
		# Get the update and add a Patchbaseline
		$Patch = get-patch | where {$_.Name -eq $PatchName}
		New-PatchBaseline -Name $BaseName -Static -IncludePatch $Patch
	}
	# Attach the baseline on the cluster
	$Cluster = Get-Cluster $Cluster
	$Baseline = Get-Baseline -Name $BaseName
	if($Baseline){
		$Baseline | Attach-Baseline -Entity $Cluster
		
		# Scan the Cluster 
		$Cluster | Scan-Inventory
		
		# Remediate the cluster with the baseline
		$Cluster | Remediate-Inventory -Baseline $Baseline -ClusterDisableDistributedPowerManagement:$true -ClusterDisableFaultTolerance:$true -ClusterDisableHighAvailability:$true -Confirm:$false -HostDisableMediaDevices:$true
		
	}
}

And here are some screenshots of it running,

One host successfully updated and the remediate migrates the VM´s and start with the other host.

Finally the hosts in the cluster are updated.

I have used some of the code from Jonathan Medd´s post on upgrading from 4.0->4.1 to create my script.