Application virtualization, IoT and Cloud Computing, Blog of Sacha Thomet


PowerManagement for dedicated Citrix desktops? Yes with Tags!

Are you using Tags in your XenApp & XenDesktop environment? Maybe you should. Tags to resources, in my case desktops can be very powerful especially in combination with PoSh scripts. You can do actions for machines depending on the tag. Of course you also can use tags to filter Citrix policies on it, also useful.

I had the Problem that I have a delivery group with dedicated Win10 desktops so for dedicated desktops there is no power management. Usually it’s also not needed because if a user launch a Citrix Session over Storefront the machine get’s powered on. The problem in my case, sometimes users connects on an other ways than Citrix to his desktop, so this built-it construct doesn’t work. So if they shutdown their virtual Desktop they can newer ever access it until an admin power it on over director or studio.

My solution to this was, I tagged this special user machines with a tag “AlwaysOnline” in Studio and I wrote this small script which runs every 15 minutes:

param([string]$tags=$(throw "Tag parameter is required"), [string]$poweroperation=$(throw "Power operaton parameter is required"))
# Created on: 09.2016 Version: 0.2
# Created by: Sacha Thomet
# File name: PowerOperation-DependingMachineTags.ps1
# Description:  This is a Powershell to change the PowerState of VDI's or XenApp Servers in
#               a PowerManaged XenDesktop 7.x environment accodring to Tags.
# Prerequisite: None, a XenDesktop Controller with according privileges necessary
# Call by : Manual  or Scheduled Task
# Load only the snap-ins, which are used
if ((Get-PSSnapin "Citrix.Broker.Admin.*" -EA silentlycontinue) -eq $null) {
try { Add-PSSnapin Citrix.Broker.Admin.* -ErrorAction Stop }
catch { write-error "Error Get-PSSnapin Citrix.Broker.Admin.* Powershell snapin"; Return }
# Change the below variables to suit your environment

$maxmachines = "1000" # as default only 250 records, this increase it to 1000
#$tags = "AlwaysOnline" # if you comment out the param line you can have the tag here
#$poweroperation = "TurnOn"  # if you comment out the param line you can have the poweroperation here

$machines = Get-BrokerMachine -MaxRecordCount $maxmachines | Where-Object {$_.tags -eq $tags }

foreach($machine in $machines)
$machinename = $machine | %{ $_.MachineName }
Write-Host "Action $poweroperation will be performed for $machinename  "
New-BrokerHostingPowerAction  -Action $poweroperation -MachineName $machinename

I know I know, this is not a common use case, but the script construct show what is possible with tags … there are almost unlimited possibility to cover special cases with tags.

My example Script on GitHub: PowerOperation-DependingMachineTags.ps1

Finally 1.0 – but never finalized!

In November 2014 I created the blog post about the adaption of the PVS Script for XenApp/XenDesktop 7.x :
XenDesktop & XenApp FMA (7.x) HealthCheck – Oops!… I Did It Again

sheepsNow after almost two years of continuous development on this Script, the XenApp & XenDesktop 7.x Health Check has now Version 1.0.

I started this script in a very basic version and in the meanwhile lot of tester and contributors helped to bring this script in this version. Just this year I started with GitHub and it’s surprising, the community who helped is awesome!

Now after a number of 0.x versions, just at the point we introduce the XML configuration file, I can say the version now has deserved the number 1.00.


The big benefit of the configuration in the XML file is in case you have multiple environments with the script you don’t need to edit the header section of the script which has earlier contained the config. On a new version of the script, you just replace the script in your environments and keep the XML file.

The XML file and the script need to be in the same directory and the Name of the XML file need to be same like the script, e.g:


The XML-Config is introduced to this script by Stefan Beckmann (Twitter: @alphasteff)


The newest version of the script you can find on GitHub: 

html script output

The HTML Output file gained some more input, even it was difficult to decide which feature request to consider and which not.

We check now CPU, Memory and disk space of controllers and workers (XenApp Server and XenDesktop VDIs). Because I learned last month’s that with 7.x and the FMA architecture it’s really possible that an environment contains number of different VDA versions, I also added this info. And for troubleshooting reason I assumed that it would be helpful to have also the hypervisor host information on this output.

The code is on GitHub:

For Bug Reports or feature Request please use GitHub, of course you can also contribute on this code!

victim of a good reputation – Low free pooled XenDesktops

The Citrix Director is not so bad, and with Version 7.7 it is still better because now it’s possible to send email alerts. But one problem I still have … our provided pooled XenDesktop which are used for remote access with private computers are victim of a good reputation … this means a good word-of-mouth advertising in our company more and more people are tempted to use it … and we need to know when the number of free pooled desktop is low. And no I won’t go the read the numbers every day in director or Citrix Studio.

So I created a small script which send me an email as soon a defined threshold of free desktops is reached:


The code is on GitHub:

Happy New Script – PVS 7.7 HealthCheck

We are close to the New Year 2016 and I want to wish you all the best for 2016!
I’m already happy now, because I can publish some hours after the PVS 7.7 release from Citrix this Script:

Today Citrix released XenApp/XenDesktop 7.7 and Citrix Provisioning Services 7.7. There are a bunch of new features, but for me the most important thing, beside of the Win10 support, is
….  drum roll ….
the new PVS PowerShell API. PoSh with PVS 7.6 and below was a pain …  To know what else is new in PVS look here: PVS 7.7 New in this release

Now PVS comes with a real PoSh interface which gave more back than a bundle of strings. I had the chance to play around with PVS 7.7 Tech Preview and so here you have already the PVS HealthCheck Script for Version 7.7 in a complete new developed version:


If you have a feature request or a bug report please post it direct on GitHub.

This version doesn’t work with PVS 7.6 and below, if you have an older version than 7.7 take this script: Citrix PVS HealthCheck 

Update 11.04.2016: Performance-Improvement, Change the order of the table, all the VDI status now at the end.
Update 09.05.2016: Bugfix & Input from Jay, get CPU/Memory/Disk usage of PVS.
Update 27.05.2016: Now my Script is on GitHub
I’m an absolute GitHub newbie … for this reason I leave it still also here as a download until I’m really comfortable with GitHub.

The code is on GitHub:

Monitor you Profile directories

Since I had worked as a Citrix administrator the size of the profile directories was always a contemporary issue. It’s a fact that with new technologies like Citrix User Profile Manager the possibilities are grown and the life is easier with exclusion lists and profile streaming features. But it’s still a topic to consider if you are a Citrix admin or system engineer. I worked in former company with tools like TreeSize ( or WinDir Stat ( But in some company they wont to spend money for tools which is only usable for one administrator or they don’t trust to OpenSource software (TreeSize).

I have the approach to solve every possible problem with on-board tools. So I created now a small script what give me the possibility to check my Profile directory with one recursive level so that I get each Size of a Profile within his subdirectories.:

# Created on: 10.2015 Version: 0.3
# Created by: Sacha Thomet
# Filename: GetSubdirSizes.ps1
# Description: This script checks some Citrix Provisioning Server, Farm, vDisk & Target device parameters.
# Prerequisite: None
# Call by : Manual
# Change Log:
# V0.1 first version
# V0.2 only 1 recursive subfolder
# V0.3 corrections of calculation

$path = "\\localhost\C$\temp\profiles\"
$top = "4000"
$outputpath = "C:\temp\SubdirSize-Report.txt"

Function Get-FolderSize
 $Sizes = 0
 ForEach ($Item in (Get-ChildItem $Path))
 If ($Item.PSIsContainer) {$Sizes = (Get-ChildItem $path"$Item" -recurse | Measure-Object -property length -sum).sum}
 Else {Write-Host "No Subfolder"}

 $SizeFormated = "{0:N2}" -f ($Sizes/1MB)
 New-Object PSObject -Property @{'Folder'=$Item;'Size'=$SizeFormated }
$outfile = Get-FolderSize $path | sort -property Size –Descending |select -first $top
$outfile | out-file $outputpath
echo $outfile

PoSh Script who alert me if I’m running out of pooled Desktops

Just a small one, last days I’ve created a small script who alert me if I’m running out of pooled Desktops.

If you run this script at regular intervals you will receive an email as soon you have less desktops free than you defined as threshold:


# Created on: 08.2015 Version: 0.2
# Created by: Sacha Thomet
# File name: Citrix-XenDesktop-Alert-low-free-desktops.ps1
# Description: Check for Free Desktops in DeliveryGroups
# Prerequisite: None
# Call by : Scheduled Task e.g every 10 minutes
# Changelog:
# V0.1 Initial Version, create report file from array FreeDesktopReport and attach this to the email.
# V0.2 Change from txt-file to formatted HTML-Mail
if ((Get-PSSnapin "Citrix.Common.Commands" -EA silentlycontinue) -eq $null) {
try { Add-PSSnapin Citrix.* -ErrorAction Stop }
catch { write-error "Error Citrix.* Powershell snapin"; Return }
# Change the below variables to suit your environment

# Variables what should be changed according your environment and wishes

$DeliveryGroups = @("Win7-Desktops","Win10-Desktops")
$minDesktops = 10
$EnvironmentName="Production XenDesktop"

# E-mail report details
$emailFrom = "[email protected]"
$emailTo = "[email protected]"
$smtpServer = ""

#=======DONT CHANGE BELOW HERE =======================================================================================

$mailbody = $mailbody + "<!DOCTYPE html>"
$mailbody = $mailbody + "<html>"

$mailbody = $mailbody + "<head>"
$mailbody = $mailbody + "<style>"
$mailbody = $mailbody + "BODY{background-color:#fbfbfb; font-family: Arial;}"
$mailbody = $mailbody + "TABLE{border-width: 1px;border-style: solid;border-color: black;border-collapse: collapse; width:60%; }"
$mailbody = $mailbody + "TH{border-width: 1px;padding: 0px;border-style: solid;border-color: black; text-align:left;}"
$mailbody = $mailbody + "TD{border-width: 1px;padding: 0px;border-style: solid;border-color: black;}"
$mailbody = $mailbody + "</style>"
$mailbody = $mailbody + "</head>"

$mailbody = $mailbody + "<body>"
$mailbody = $mailbody + "This is the Low-Desktop-Alert for $EnvironmentName, if you receive this mail the value of free desktops is below the configured threshold of $minDesktops desktops! <br><br>"

$FreeDesktopReport = @()

foreach($dg in $DeliveryGroups)
$desktops = Get-BrokerDesktopGroup | where {$_.Name -eq $dg }
$CurrentDeliveryGroup = "" | Select-Object Name, Alert, DesktopsAvailable

# Write Array Values
$CurrentDeliveryGroup.Name = $dg

$CurrentDeliveryGroup.DesktopsAvailable = $desktops.DesktopsAvailable

if ($desktops.DesktopsAvailable -lt $minDesktops )
Write-Host "Number of free desktops to low for DeliveryGroup $dg, sending email"
# Add Line to Report
$CurrentDeliveryGroup.alert = "True"

$FreeDesktopReport += $CurrentDeliveryGroup

$mailbody += $FreeDesktopReport | ConvertTo-Html
$mailbody += "<br><br>Launch Citrix Studio or browse to <a href=$directoraddress>Citrix Director</a> see more information about the current Desktop usage<br>"

$mailbody = $mailbody + "<body>"
$mailbody = $mailbody + "</html>"

# If any record raises an alert, send an email.
if (($FreeDesktopReport | where {$_.alert -eq "True"}) -ne $null) {Send-MailMessage -to $emailTo -from $emailFrom -subject "********* Low free Desktop Alert for $EnvironmentName *********" -Body $mailbody -BodyAsHtml -SmtpServer $smtpServer }

Or download the Script here

GeekOvation – PVS Script on Top3

geek-zone-150x150My Citrix PVS Health Check PowerShell Script was nominated into the final round (Top3) of the Citrix GeekOvation Award. This means I have to present my “geeky solution” on Citrix Synergy 2015 in Orlando on the SYN501 Geek Speak Tonight Session. To be honest to speak in front of so much people is a new thing for me … and it makes me “a bit” twitchy …

To be nominated for this Award was not just a big surprise for me, I see this also as a honor and glory for my work. I would say it’s not so a big thing this script and it isn’t rocket science.

Dear GeekSpeak Tonight audience, please check out my PVS Health Check Script and provide me feedback if you want. Also have a look to my XenApp / XenDesktop Script

At this place  I want to say thank you to Jason Poyner who gave me the idea for such kind of Script with his XenApp6 HealthCheck and where I took some code lines for the HTML output. To Martin Hartmann who helped me by learning PowerShell and of course to Aaron Parker ( who was  the assigned CTP helping me polish off my presentation materials.

Update 14.5.2015:

I won – thank YOU

geekovationI won the competition and seems that I’m the not-yet-known Geek #1 this Year! Thanks you all! When I saw the work from to other two guys, I had the feeling to have no chance, not just because I’m the only not-native english speaker. David Ott with a great PowerShell script to modify the user profile and Shane O’Neil with the C-Sharp code to have a self healing VDI, GREAT!

Read this blog post from Shane and Paul, if you are interested on more PVS PowerShell stuff: and watch the recording of “SYN514: Turn XenApp and XenDesktop into capacity on demand with Provisioning Services automation”

I’m sure I also had luck to win, and now I have the chance to go to Las Vegas Citrix Synergy 2016. I also want to thank ControlUp for the AppleWatch they sponsored for the GeekOvation Winner.

About the code

Like already mentioned the idea about the health check with the HTML output comes from Jason Poyner. The main script with all the logics it’s from me. The hardest thing when you Script with PoSh on PVS that it doesn’t return proper PS Object, that mean’s you doing a lot of format-cut stuff to get what you want. If you want to know more about that read Paul’s blog section “PVS comes with a PowerShell add-in that is not an easy beast to tame.

What’s next?

I’m currently working on a Script which shows PVS retries during a variable timeframe (day, week or month) and create a graphic output, in this way you can see at which time you maybe have a traffic jam on your network.

I also made a script to export/import target’s into PVS and another one for export and import in XenDesktop, if I have time I’ll polish and publish this here.

XenDesktop & XenApp FMA (7.x) HealthCheck – Oops!… I Did It Again

Some months ago I’ve created the Citrix PVS Health-Check Script which is a based on the idea and some parts of code from the Health-Check Script for XenApp 6.x of Jason Poyer ( .
Because now XenApp 7 with the Release 7.6 is finally in a state where considering an upgrade of the 6.x farms make sense, I belief that the demand for a XenApp 7.x Health Check Script grows.

So I did it again and took the “HealthCheck framework” to build a new version which combines the Power of the Citrix PS-Snappins for XenDesktop/XenApp and  the HTML-Output-Script of the existing HealthCheck Scripts.

The result is a new HealthCheck Script which is usable for XenApp and XenDesktop 7.x and what makes me also happy, with only a few line of more code the Script is downwardly compatible for XenDesktop 5.6 environments.

  XenDesktop XenApp Health Check HTML Output

This is just the first version and I’m sure that more check’s need to be added. Feedback and “Feature requests” are welcome … And to be honest I have not yet a big environment to test my Script, so please be insightfully if you find some bugs and report them to me.

In the first part of the Script you are able to configure some parameters. You can decide if you only want to see the “bad” Desktops on which something it’s going wrong or if you want see everything. In huge XenDesktop environments you want definitely only see the bad machines …  ( $ShowOnlyErrorVDI = 1 ) Also you can decide if you want only report XenApp or only XenDesktop or both. The Desktops and XenApps are in two different Tables. It’s also possible to exclude Collections ($ExcludeCatalogs) from the Check, so virtual Desktops which are for testing purposes are not checked.



If you have a feature request or a bug report please post it direct on GitHub.

Update 12.05.2016 (Version 0.95):
– Check CPU, Memory and C: of Controllers
– XenApp: Add values: CPU & Memory and Disk Usage
– XenApp: Option to toggle on/off to show Connected Users
– XenApp: DesktopFree set to N/A because not relevant
If you need a Health Check Script for XenApp Version which are older than XenApp 7.x see where it’s an excellent work and the inspiration for my HealthCheck-Scripts!

The code is on GitHub:

For Bug Reports or feature Request please use GitHub, of course you can also contribute on this code!

Citrix PVS HealthCheck

Problem: There is currently no free tool to check the health status of a Citrix Provisioning Services farm with all Citrix relevant parameters. There are monitoring tools which tell you if a server is up and running or if it’s not, but now tool who gives you a special overview for Citrix Provisioning Services according to how much targets on which server, how much communication retries on which target an so on.

Since some year’s I’m using the XenApp Health Check from Jason Poyner ( to get daily a status of my environment(s). I really appreciate the work from Jason, if you don’t know it until now, you really need have a look on his blog and on the Script . The mentioned XenApp Farm Health Check Script gave me the inspiration to create a similar Health Check Script for my Citrix Provisioning Server farm. To be honest, and I’m happy that Jason doesn’t mind, that I have “recycled” some parts, or say the “framework” of his XenApp Farm Health Check script to build my Script.

The script I built can be scheduled with a Windows task which trigger a run once a day and send out a email with the attached HTML-Report:

I’ve written the Script to perform the checks in 3 Parts: 

Target Devices:

  • Ping
  • Retry (Threshold configurable, default is 15)
  • PVS Disk (Value which I read from PVS farm with get-mcli)
  • PVS Disk (Value which is written in the personality.ini on the Target)
  • PVS disk version (V1.2)
  • The vDisk-Store *
  • PVS Server
  • WriteCache Size

* like you can read in my BlogPost “Achieve fastest Citrix Provisioning Target Device” I propose to have the vDisk Store local on each provisioning server. So I can check with that column if the Target is booting from the correct vDisk store. The Master Store is on CIFS and only used for updates of the Image.


  • Replication state (PVS don’t replicate anything – but can check if the vDisks on all servers are equal)
  • DeviceCount to see how many targets are currently reading from this Image
  • Load Balancing algorithm
    (Green if Best Effort because I want to have subnet affinity – feel free to change for your need.  I have two data centers with two separated subnet’s – so I can guarantee that each target stream from his own data centre as long the server is available. Just in case of an outage of all PVS server in one DC it fail over to the other DC  )
  • WriteCache Type
  • Name & Date of the vDisk

PVS Servers:

  • Ping
  • Active
  • How many device are connected

Don’t forget to register the PVS SnapIn DLL e.g. with .Net Framework 2.0 with the command:

“C:\WINDOWS\Microsoft.NET\Framework64\v2.0.50727\InstallUtil.exe” “C:\Program Files\Citrix\Provisioning Services Console\McliPSSnapIn.dll”

Otherwise you get a blank output. If you have another .Net Framework version or path, please adjust the path in the command! 

# Created on: 08.2014            Version: 1.63
# Created by: Sacha Thomet, /
# Filename: Citrix-PVS-Farm-Health-toHTML.ps1
# Special Thanks to:
# - Jason Poyner ... I've borrowed parts of the script and ideas to create this 
#   PVS health check script. Check his excellent XenApp Health Check @
# - Martin Hartmann to share his PowerShell KnowHow with me.
# Description: This script checks Citrix Provisioning Server, Farm, vDisk & Target devices.
# Prerequisite: Script must run on a PVS server, where MCLI snap-in is registered.
# Register SnapIn with command: C:\WINDOWS\Microsoft.NET\Framework64\v4.0.30319\InstallUtil.exe 
# "C:\Program Files\Citrix\Provisioning Services Console\McliPSSnapIn.dll"
# Call by : Scheduled Task, e.g. once a day
# Change Log: 
#      V1.1:  Consolidated code
#      V1.2:  Add possibility to check only specified Versions
#      V1.3:  New version after Citrix Synergy 2015 GeekOvation Award nomination. 
#		      With correction of typos and add some documentation lines. 
#      V1.4:  Possibility for multiple stores (Thanks to Kafedzhiev  for the code) (08-2015) 
#      V1.5:  Show create date of vDisk, FileName and the count of the used vDisk (09-2015)
#      V1.6:  Add RamCache used from each target, added code by Jonathan Pitre, 
#		      code from Matthew Nics (10-2015) 
#      V1.61: Changed RamCache to general WriteCache, add possibility to get Size of Cache on HD
#      V1.62: Correction in Header to show correct farm name instead a "6", correction in 
#		      LoadBalancingAlgorithm, Error if a disk is assigned fix to a server. 
#      V1.63: Check of Stream-, Soap-, and TFTP-Service (12-2015)
if ((Get-PSSnapin "McliPSSnapIn" -EA silentlycontinue) -eq $null) {
try { Add-PSSnapin McliPSSnapIn -ErrorAction Stop }
catch { write-error "Error loading PVS McliPSSnapIn PowerShell snapin"; Return }
# Change the below variables to suit your environment
# Target Device Health Check threshold:
$retrythresholdWarning= "15" # define the Threshold from how many retries the color switch to red
# Include for Device Collections, type "every" if you want to see every Collection 
# Example1: $Collections = @("XA65","XA5")
# Example2: $Collections = @("every")
$Collections = @("every")
# Information about the site you want to check:
$siteName="site" # site name on which the according Store is.
# E-mail report details
$emailFrom = "[email protected]"
$emailTo = "[email protected]"#,"[email protected]"
$smtpServer = ""
$emailSubjectStart = "PVS Farm Report"
$mailprio = "High"
$currentDir = Split-Path $MyInvocation.MyCommand.Path
$logfile = Join-Path $currentDir ("PVSHealthCheck.log")
$resultsHTM = Join-Path $currentDir ("PVSFarmReport.htm")
$errorsHTM = Join-Path $currentDir ("PVSHealthCheckErrors.htm") 
#Header for Table 1 "Target Device Checks"
$TargetfirstheaderName = "TargetDeviceName"
$TargetheaderNames = "CollectionName", "Ping", "Retry", "vDisk_PVS", "vDisk_Version", "WriteCache", "PVSServer"
$TargetheaderWidths = "4", "4", "4", "4", "2" , "4", "4"
$Targettablewidth = 1200
#Header for Table 2 "vDisk Checks"
$vDiksFirstheaderName = "vDisk"
$vDiskheaderNames = "Store", "vDiskFileName", "deviceCount", "CreateDate" , "ReplState", "LoadBalancingAlgorithm", "WriteCacheType"
$vDiskheaderWidths = "4", "8", "2","4", "4", "4", "4"
$vDisktablewidth = 1200
#Header for Table 3 "PV Server"
$PVSfirstheaderName = "PVS Server"
$PVSHeaderNames = "Ping", "Active", "deviceCount","SoapService","StreamService","TFTPService"
$PVSheaderWidths = "4", "4", "4","4","4","4"
$PVStablewidth = 600
#Header for Table 4 "Farm"
$PVSFirstFarmheaderName = "FarmChecks"
$PVSFarmHeaderNames = "Setting", "Value"
$PVSFarmWidths = "4", "8", "8"
$PVSFarmTablewidth = 400
#log function
function LogMe() {
[parameter(Mandatory = $true, ValueFromPipeline = $true)] $logEntry,
 if ($error) {
$logEntry = "[ERROR] $logEntry" ; Write-Host "$logEntry" -Foregroundcolor Red}
elseif ($warning) {
Write-Warning "$logEntry" ; $logEntry = "[WARNING] $logEntry"}
elseif ($progress) {
Write-Host "$logEntry" -Foregroundcolor Green}
elseif ($display) {
Write-Host "$logEntry" }
 #$logEntry = ((Get-Date -uformat "%D %T") + " - " + $logEntry)
$logEntry | Out-File $logFile -Append
function Ping([string]$hostname, [int]$timeout = 200) {
$ping = new-object System.Net.NetworkInformation.Ping #creates a ping object
try {
$result = $ping.send($hostname, $timeout).Status.ToString()
} catch {
$result = "Failure"
return $result
Function writeHtmlHeader
param($title, $fileName)
$date = ( Get-Date -format R)
$head = @"
<meta http-equiv='Content-Type' content='text/html; charset=iso-8859-1'>

<STYLE TYPE="text/css">
<!-- td { font-family: Tahoma; font-size: 11px; border-top: 1px solid #999999; border-right: 1px solid #999999; border-bottom: 1px solid #999999; border-left: 1px solid #999999; padding-top: 0px; padding-right: 0px; padding-bottom: 0px; padding-left: 0px; overflow: hidden; } body { margin-left: 5px; margin-top: 5px; margin-right: 0px; margin-bottom: 10px; table { table-layout:fixed; border: thin solid #000000; } -->


<table width='1200'>

<tr bgcolor='#CCCCCC'>

<td colspan='7' height='48' align='center' valign="middle">
<font face='tahoma' color='#003399' size='4'>
<strong>$title - $date</strong></font>



$head | Out-File $fileName
# ==============================================================================================
Function writeTableHeader
param($fileName, $firstheaderName, $headerNames, $headerWidths, $tablewidth)
$tableHeader = @"

<table width='$tablewidth'>


<tr bgcolor=#CCCCCC>

<td width='6%' align='center'><strong>$firstheaderName</strong></td>

$i = 0
while ($i -lt $headerNames.count) {
$headerName = $headerNames[$i]
$headerWidth = $headerWidths[$i]
$tableHeader += "

<td width='" + $headerWidth + "%' align='center'><strong>$headerName</strong></td>

$tableHeader += "</tr>

$tableHeader | Out-File $fileName -append
# ==============================================================================================
Function writeTableFooter

"| Out-File $fileName -append
Function writeData
param($data, $fileName, $headerNames)
 $data.Keys | sort | foreach {
$tableEntry += "

$computerName = $_
$tableEntry += ("

<td bgcolor='#CCCCCC' align=center><font color='#003399'>$computerName</font></td>

#$data.$_.Keys | foreach {
$headerNames | foreach {
#"$computerName : $_" | LogMe -display
try {
if ($data.$computerName.$_[0] -eq "SUCCESS") { $bgcolor = "#387C44"; $fontColor = "#FFFFFF" }
elseif ($data.$computerName.$_[0] -eq "WARNING") { $bgcolor = "#FF7700"; $fontColor = "#FFFFFF" }
elseif ($data.$computerName.$_[0] -eq "ERROR") { $bgcolor = "#FF0000"; $fontColor = "#FFFFFF" }
else { $bgcolor = "#CCCCCC"; $fontColor = "#003399" }
$testResult = $data.$computerName.$_[1]
catch {
$bgcolor = "#CCCCCC"; $fontColor = "#003399"
$testResult = ""
 $tableEntry += ("

<td bgcolor='" + $bgcolor + "' align=center><font color='" + $fontColor + "'>$testResult</font></td>

 $tableEntry += "</tr>

 $tableEntry | Out-File $fileName -append
# ==============================================================================================
Function writeHtmlFooter


<table width='1200'>

<tr bgcolor='#CCCCCC'>

<td colspan='7' height='25' align='left'>

<font face='courier' color='#000000' size='2'><strong>Retry Threshold =</strong></font><font color='#003399' face='courier' size='2'> $retrythresholdWarning


<tr bgcolor='#CCCCCC'>


<tr bgcolor='#CCCCCC'>


"@ | Out-File $FileName -append
rm $logfile -force -EA SilentlyContinue
"Begin with Citrix Provisioning Services HealthCheck" | LogMe -display -progress
" " | LogMe -display -progress
# ======= PVS Target Device Check ========
"Check PVS Target Devices" | LogMe -display -progress
" " | LogMe -display -progress
$allResults = @{}
$pvsdevices = mcli-get device -f deviceName | Select-String deviceName
foreach($target in $pvsdevices)
 $tests = @{} 
 # Check to see if the server is in an excluded folder path
$target | Select-String deviceName 
 $_targetshort = $target -replace "deviceName: ",""
 $pvcollectionName = mcli-get deviceinfo -p devicename=$_targetshort | select-string collectionName
$short_collectionName = $pvcollectionName.ToString().TrimStart("collectionName: ")
 #Only Check Servers in defined Collections: 
 if ($Collections -contains $short_collectionName -Or $Collections -contains "every") { 
 $target | Select-String deviceName 
 $_targetshort = $target -replace "deviceName: ",""
$_targetshort | LogMe -display -progress
 # Ping server 
 $result = Ping $_targetshort 100
if ($result -ne "SUCCESS") { $tests.Ping = "ERROR", $result }
else { $tests.Ping = "SUCCESS", $result 
$pvcollectionName = mcli-get deviceinfo -p devicename=$_targetshort | select-string collectionName
$short_collectionName = $pvcollectionName.ToString().TrimStart("collectionName: ")
$tests.CollectionName = "NEUTRAL", "$short_collectionName"
 # Test Retries
$devicestatus = mcli-get deviceinfo -p devicename=$_targetshort -f status
$retrycount = $devicestatus[4].TrimStart("status: ") -as [int]
if ($retrycount -lt $retrythresholdWarning) { $tests.Retry = "SUCCESS", "$retrycount Retry = OK" }
else { $tests.Retry = "WARNING","$retrycount retries!" }
 #Check assigned Image
$devicediskFileName = mcli-get deviceinfo -p devicename=$_targetshort | select-string diskFileName
$short_devicediskFileName = $devicediskFileName.ToString().TrimStart("diskFileName: ")
$tests.vDisk_PVS = "SUCCESS", "$short_devicediskFileName"
 #Check assigned Image Version
$devicediskVersion = mcli-get deviceinfo -p devicename=$_targetshort | select-string diskVersion:
$short_devicediskVersion = $devicediskVersion.ToString().TrimStart("diskVersion: ")
$tests.vDisk_Version = "SUCCESS", "$short_devicediskVersion"
$PVSServername = mcli-get deviceinfo -p devicename=$_targetshort | select-string serverName
$short_PVSServername = $PVSServername.ToString().TrimStart("serverName: ")
$tests.PVSServer = "Neutral", "$short_PVSServername"

################ PVS WriteCache SECTION ###############

		if (test-path \\$_targetshort\c$\Personality.ini)

			$wconhd = ""
			$wconhd = Get-Content \\$_targetshort\c$\Personality.ini | Where-Object  {$_.Contains("WriteCacheType=4") }
			If ($wconhd -match "$WriteCacheType=4") {Write-Host Cache on HDD
			#WWC on HD is $wconhd

				# Relative path to the PVS vDisk write cache file
				$PvsWriteCache   = "d$\.vdiskcache"
				# Size of the local PVS write cache drive
				$PvsWriteMaxSize = 10gb # size in GB
				$PvsWriteCacheUNC = Join-Path "\\$_targetshort" $PvsWriteCache 
				$CacheDiskexists  = Test-Path $PvsWriteCacheUNC
				if ($CacheDiskexists -eq $True)
					$CacheDisk = [long] ((get-childitem $PvsWriteCacheUNC -force).length)
					$CacheDiskGB = "{0:n2}GB" -f($CacheDisk / 1GB)
					"PVS Cache file size: {0:n2}GB" -f($CacheDisk / 1GB) | LogMe
					#"PVS Cache max size: {0:n2}GB" -f($PvsWriteMaxSize / 1GB) | LogMe -display
					if($CacheDisk -lt ($PvsWriteMaxSize * 0.5))
					   "WriteCache file size is low" | LogMe
					   $tests.WriteCache = "SUCCESS", $CacheDiskGB
					elseif($CacheDisk -lt ($PvsWriteMaxSize * 0.8))
					   "WriteCache file size moderate" | LogMe -display -warning
					   $tests.WriteCache = "WARNING", $CacheDiskGB
					   "WriteCache file size is high" | LogMe -display -error
					   $tests.WriteCache = "ERORR", $CacheDiskGB
				$Cachedisk = 0
				$VDISKImage = get-content \\$_targetshort\c$\Personality.ini | Select-String "Diskname" | Out-String | % { $_.substring(12)}
				if($VDISKImage -Match $DefaultVDISK){
					"Default vDisk detected" | LogMe
					$tests.vDisk = "SUCCESS", $VDISKImage
				} else {
					"vDisk unknown"  | LogMe -display -error
					$tests.vDisk = "SUCCESS", $VDISKImage
			{Write-Host Cache on Ram
			#Get-RamCache from each target, code from Matthew Nics
			$RAMCache = [math]::truncate((Get-WmiObject Win32_PerfFormattedData_PerfOS_Memory -ComputerName $_targetshort).PoolNonPagedBytes /1MB)
			$tests.WriteCache = "Neutral", "$RamCache MB on Ram"
		{Write-Host WriteCache not readable
		$tests.WriteCache = "Neutral", "Cache not readable"	
		############## END PVS WriteCache SECTION #############

#Forward results to $allResult array which will be written in HTM-File
$allResults.$_targetshort = $tests
# ======= PVS vDisk Check #==================================================================
"Check PVS vDisks" | LogMe -display -progress
" " | LogMe -display -progress
$storenames = mcli-get store | Select-string storename
$vdiskResults = @{}
foreach ($storenameA in $storenames)
$storename = $storenameA -replace "storename: ",""
$storeid = mcli-get store -p storeName=$storename | Select-String storeId
$storeid_short = $storeid -replace "storeId: ",""
$alldisks = Mcli-Get disklocator -p siteName=$siteName, storeId=$storeid_short | Select-String diskLocatorName
foreach($disk in $alldisks)
$disk1 = $disk | Select-String diskLocatorName
$disklocator_short = $disk1 -replace "diskLocatorName: ",""
foreach($diksloc in $disklocator_short)
 $VDtests = @{} 
 $DiskVersion = Mcli-Get DiskVersion -p diskLocatorName=$disklocator_short, siteName=$siteName, storeName=$storename
$diskreplstatus = $DiskVersion | Select-String goodInventoryStatus
$diskreplstatus_short = $diskreplstatus -replace "goodInventoryStatus: ","" 
 # vDiskFileName & createDate 
 $pathA = mcli-get store -p storeName=$storename | Select-String path -casesensitive
$path = $pathA -replace "path: ",""
 $diskfilenameA = Mcli-Get DiskVersion -p diskLocatorName=$disklocator_short, siteName=$siteName, storeName=$storename | Select-String diskFileName 
 $diskfilename = $diskfilenameA -replace "diskFileName: ","
 $createDateA = Mcli-Get DiskVersion -p diskLocatorName=$disklocator_short, siteName=$siteName, storeName=$storename | Select-String createDate 
 $createDate = $createDateA -replace "createDate: ","
 $VDtests.vDiskFileName = "OK", " $diskfilename"
Write-Host ("Path is $path $disklocator_short $diskfilename")
 $VDtests.createDate = "OK", " $createDate"
Write-Host ("Path is $path $disklocator_short $createDate")
 $vdiskResults.$disklocator_short = $VDtests
 #Check if correct replicated
if($diskreplstatus_short -eq 1 ){
"$disklocator_short correct replicated" | LogMe
$VDtests.ReplState = "SUCCESS", "Replication is OK"
 } else {
"$disklocator_short not correct replicated " | LogMe -display -error
$VDtests.ReplState = "ERROR", "Replication is NOT OK"
 # Check deviceCount: 
 $diskdevicecount = $DiskVersion | Select-String deviceCount
$diskdevicecounts_short = $diskdevicecount -replace "deviceCount: ","
 $VDtests.deviceCount = "OK", "$diskdevicecounts_short "
 #Label Storename 
 $VDtests.Store = "OK", " $storename "
Write-Host ("Store is $storename")
 $vdiskResults.$disklocator_short = $VDtests
# Check for LB-Algorithm
# ----------------------
# Feel free to change it to the the from you desired State (e.g.Exchange a SUCCESS with a WARNING)
# In this default configuration "BestEffort" or "None" is desired and appears green on the output.
# is desired)

#ServeName must be empty! otherwise no LB is active!
$LBnoServer = ""
$LBnoServer_short = ""
$LBnoServer = Mcli-Get disklocator -p siteName=$siteName, storeName=$storename, diskLocatorName=$disklocator_short | Select-String serverName
$LBnoServer_short = $LBnoServer -replace "serverName: ","" 
Write-Host ("vDisk is fix assigned to $LBnoServer")
#not assigned to a server
if ($LBnoServer_short -eq "")
		$LBAlgo = Mcli-Get disklocator -p siteName=$siteName, storeName=$storename | Select-String subnetAffinity
		$LBAlgo_short = $LBAlgo -replace "subnetAffinity: ","" 
		#SubnetAffinity: 1=Best Effort, 2= fixed, 0=none
		if($LBAlgo_short -eq 1 ){
		"LB-Algorythm is set to BestEffort" | LogMe
		$VDtests.LoadBalancingAlgorithm = "SUCCESS", "LB is set to BEST EFFORT"} 
		 elseif($LBAlgo_short -eq 2 ){
		"LB-Algorythm is set to fixed" | LogMe
		$VDtests.LoadBalancingAlgorithm = "WARNING", "LB is set to FIXED"}
		 elseif($LBAlgo_short -eq 0 ){
		"LB-Algorythm is set to none" | LogMe
		$VDtests.LoadBalancingAlgorithm = "SUCCESS", "LB is set to NONE, least busy server is used"}


#Disk fix assigned to a server
$VDtests.LoadBalancingAlgorithm = "ERROR", "vDisk is fix assigned to $LBnoServer, no LoadBalancing!"}
 #Check for WriteCacheType
# -----------------------
# Feel free to change it to the the from you desired State (e.g.Exchange a SUCCESS with a WARNING)
# In this default configuration, only "Cache to Ram with overflow" and "Cache to Device Hard disk" is desired and appears green on the output.
 $WriteCacheType = Mcli-Get DiskInfo -p diskLocatorName=$disklocator_short, siteName=$siteName, storeName=$storename
$WriteCacheType_short = $WriteCacheType -replace "WriteCacheType: ",""
 #$WriteCacheType 9=RamOfToHD 0=PrivateMode 4=DeviceHD 8=DeviceHDPersistent 3=DeviceRAM 1=PVSServer 7=ServerPersistent 
 if($WriteCacheType_short -eq 9 ){
"WC is set to Cache to Device Ram with overflow to HD" | LogMe
$VDtests.WriteCacheType = "SUCCESS", "WC Cache to Ram with overflow to HD"}
 elseif($WriteCacheType_short -eq 0 ){
"WC is not set because vDisk is in PrivateMode (R/W)" | LogMe
$VDtests.WriteCacheType = "Error", "vDisk is in PrivateMode (R/W) "}
 elseif($WriteCacheType_short -eq 4 ){
"WC is set to Cache to Device Hard Disk" | LogMe
$VDtests.WriteCacheType = "SUCCESS", "WC is set to Cache to Device Hard Disk"}
 elseif($WriteCacheType_short -eq 8 ){
"WC is set to Cache to Device Hard Disk Persistent" | LogMe
$VDtests.WriteCacheType = "Error", "WC is set to Cache to Device Hard Disk Persistent"}
 elseif($WriteCacheType_short -eq 3 ){
"WC is set to Cache to Device Ram" | LogMe
$VDtests.WriteCacheType = "WARNING", "WC is set to Cache to Device Ram"}
 elseif($WriteCacheType_short -eq 1 ){
"WC is set to Cache to PVS Server HD" | LogMe
$VDtests.WriteCacheType = "Error", "WC is set to Cache to PVS Server HD"}
 elseif($WriteCacheType_short -eq 7 ){
"WC is set to Cache to PVS Server HD Persistent" | LogMe
$VDtests.WriteCacheType = "Error", "WC is set to Cache to PVS Server HD Persistent"}

# ======= PVS Server Check ==================================================================
"Check PVS Servers" | LogMe -display -progress
" " | LogMe -display -progress
$PVSResults = @{}
$allPVSServer = mcli-get server | Select-String serverName
foreach($PVServerName in $allPVSServer)
$PVStests = @{} 
 $PVServerName1 = $PVServerName | Select-String serverName
$PVServerName_short = $PVServerName1 -replace "serverName: ","" 
 # Ping server 
 $result = Ping $PVServerName_short 100
if ($result -ne "SUCCESS") { $PVStests.Ping = "ERROR", $result }
else { $PVStests.Ping = "SUCCESS", $result 
 #Check PVS Service Status
$serverstatus = mcli-get ServerStatus -p serverName=$PVServerName_short -f status
$actviestatus = $serverstatus[4].TrimStart("status: ") -as [int]
if ($actviestatus -eq 1) { $PVStests.Active = "SUCCESS", "active" }
else { $PVStests.Active = "Error","inactive" }

# Check services
		if ((Get-Service -Name "soapserver" -ComputerName $PVServerName_short).Status -Match "Running") {
			"SoapService running..." | LogMe
			$PVStests.SoapService = "SUCCESS", "Success"
		} else {
			"SoapService service stopped"  | LogMe -display -error
			$PVStests.SoapService = "ERROR", "Error"
		if ((Get-Service -Name "StreamService" -ComputerName $PVServerName_short).Status -Match "Running") {
			"StreamService service running..." | LogMe
			$PVStests.StreamService = "SUCCESS","Success"
		} else {
			"StreamService service stopped"  | LogMe -display -error
			$PVStests.StreamService = "ERROR","Error"
		if ((Get-Service -Name "BNTFTP" -ComputerName $PVServerName_short).Status -Match "Running") {
			"TFTP service running..." | LogMe
			$PVStests.TFTPService = "SUCCESS","Success"
		} else {
			"TFTP  service stopped"  | LogMe -display -error
			$PVStests.TFTPService = "ERROR","Error"
 #Check PVS deviceCount
$serverdevicecount = mcli-get ServerStatus -p serverName=$PVServerName_short -f deviceCount
$numberofdevices = $serverdevicecount[4].TrimStart("deviceCount: ") -as [int]
if ($numberofdevices -gt 1) { $PVStests.deviceCount = "SUCCESS", " $numberofdevices active" }
else { $PVStests.deviceCount = "WARNING","No devices on this server" }
 $PVSResults.$PVServerName_short = $PVStests
# ======= PVS Farm Check ====================================================================
"Read some PVS Farm Parameters" | LogMe -display -progress
" " | LogMe -display -progress
$PVSFarmResults = @{}
$PVSfarms = mcli-get Farm #| Select-String FarmName

$farmname = mcli-get Farm | Select-String FarmName
$farmname_short = $farmname -replace "farmName: ",""

foreach($PVSFarm in $PVSfarms)
$PVSFarmtests = @{}
# remove not needed record parts
if ($PVSFarm -like '*description*'){continue;}
if ($PVSFarm -like '*record*'){continue;}
if ($PVSFarm -like '*failover*'){continue;}
if ($PVSFarm -like '*executing*'){continue;}
if ($PVSFarm -like '*defaultSiteName*'){continue;}
if ($PVSFarm -like '*autoAddEnabled*'){continue;}
if ($PVSFarm -like '*role*'){continue;}
if ($PVSFarm -like '*audit*'){continue;}
if ($PVSFarm -like '*defaultSiteId*'){continue;}
if ($PVSFarm -like '*maxVersions*'){continue;}
if ($PVSFarm -like '*databaseInstanceName*'){continue;}
if ($PVSFarm -like '*farmId*'){continue;}
if ($PVSFarm -like '*merge*'){continue;}
if ($PVSFarm -like '*adGroups*'){continue;}
 if ($PVSFarm -ne '') {
$Nr += 1
$arr = $PVSFarm -split ': '
$farmsetting = $arr[0]
$PVSFarmtests.Setting = "NEUTRAL", "$farmsetting"
$arr = $PVSFarm -split ': '
$farmsettingvalue = $arr[1]
$PVSFarmtests.Value = "NEUTRAL", "$farmsettingvalue"
$PVSFarmResults.$farmnr = $PVSFarmtests
# ======= Write all results to an html file =================================================
Write-Host ("Saving results to html report: " + $resultsHTM)
writeHtmlHeader "PVS Farm Report $farmname_short" $resultsHTM
writeTableHeader $resultsHTM $TargetFirstheaderName $TargetheaderNames $TargetheaderWidths $TargetTablewidth
$allResults | sort-object -property collectionName | % { writeData $allResults $resultsHTM $TargetheaderNames}
writeTableFooter $resultsHTM
writeTableHeader $resultsHTM $vDiksFirstheaderName $vDiskheaderNames $vDiskheaderWidths $vDisktablewidth
$vdiskResults | sort-object -property ReplState | % { writeData $vdiskResults $resultsHTM $vDiskheaderNames }
writeTableFooter $resultsHTM
writeTableHeader $resultsHTM $PVSFirstheaderName $PVSheaderNames $PVSheaderWidths $PVStablewidth
$PVSResults | sort-object -property PVServerName_short | % { writeData $PVSResults $resultsHTM $PVSheaderNames}
writeTableFooter $resultsHTM
writeTableHeader $resultsHTM $PVSFirstFarmheaderName $PVSFarmHeaderNames $PVSFarmWidths $PVSFarmTablewidth
$PVSFarmResults | % { writeData $PVSFarmResults $resultsHTM $PVSFarmHeaderNames}
writeTableFooter $resultsHTM
writeHtmlFooter $resultsHTM
#send email
$emailSubject = ("$emailSubjectStart - $farmname_short - " + (Get-Date -format R))
$mailMessageParameters = @{
From = $emailFrom
To = $emailTo
Subject = $emailSubject
SmtpServer = $smtpServer
Body = (gc $resultsHTM) | Out-String
Attachment = $resultsHTM
# Send mail if you wish
Send-MailMessage @mailMessageParameters -BodyAsHtml -Priority $mailprio

or Download the file on GitHub: Citrix-PVS-Farm-Health-toHTML.ps1

Question or suggestion for improvement? Just contact me.

Update 12.11.14: Now in version 1.2 it’s possible to define the device collection which should be checked. An other additional feature is that the script shows the version of the vDisk.
Update 15.08.15: Now in version 1.4 it’s possible to have more than just one vDisk store.
Update 16.09.15: Now in version 1.5 the vDisk file name, it’s date and the count of the used disk will be showed.
Update 14.10.15: Version 1.61 New feature, PVS Cache (on disk or Ram) will be reported. / 20.10.2015 1.62: fixed some bugs (see comments)

Update 29.12.15: If you use PVS 7.7 or higher check out the new blog post with the 7.7 Script:

Because of the Release of PVS with the new PoSh API the further development of this Script is discontinued. Sorry.

Create published applications XenApp to each server

For testing purposes it can be helpful to have a published application without loadbalancing to every single server of your Citrix Farm. The creation of such application can be easily done with the following example script which create a CMD on every farm server:

Add-PSSnapin Citrix.XenApp.Commands -ErrorAction SilentlyContinue
$serverlist = Get-XAServer #Get all server from farm

# loop through all servers

 foreach($srv in $serverlist)

 echo $srv.ServerName

 New-XAApplication -BrowserName "cmd on $srv" -ApplicationType "ServerInstalled" -DisplayName "cmd on $srv" -FolderPath "Applications/Published Apps/z_admin" -ClientFolder "Administration\cmd" -Enabled $true -CommandLineExecutable "C:\Windows\system32\cmd.exe" -WorkingDirectory "C:\Windows\system32" -AnonymousConnectionsAllowed $false -AddToClientStartMenu $false -InstanceLimit "-1" #-WindowType "1024×768" -ColorDepth "Colors256"

 Add-XAApplicationAccount -BrowserName "cmd on $srv" -Accounts "domain\USER1"
 Add-XAApplicationAccount -BrowserName "cmd on $srv" -Accounts "domain\USER2"
 Add-XAApplicationServer -BrowserName "cmd on $srv" -ServerNames $srv.ServerName


By the way … if you made a mistake you can remove the created applications by the same foreach-loop with Remove-XAApplication:


$serverlist = Get-XAServer #Get all server from farm
# loop through all servers
foreach($srv in $serverlist)
 echo $srv.ServerName
 Remove-XAApplication -BrowserName "cmd on $srv"


All scripts are provided AS IS without warranty of any kind.