In One Liner: Configuring Shutdown Tracker in Windows Server I mentioned that it’s often preferable to quickly configure some server settings when building servers. As a consultant, I like to set up my server profile when building servers in a manner that’s efficient and convenient for me. One thing that drives me completely insane is the default taskbar group setting. Taskbar grouping is how Windows groups common items together on the taskbar. By default, all similar items are lumped together, i.e. all Internet Explorer windows. So to go back to an IE window could take two mouse clicks instead of one. Let’s take a look at streamlining this configuration for Server 2012 and Server 2012 R2.

Taskbar grouping has three settings. The default “always combine” mentioned previously, “combine when taskbar full” which doesn’t start grouping until there are enough items to fill the taskbar, and my favorite, “never combine”. As you can probably guess, “never combine” doesn’t group taskbar items at all. Since I usually don’t have more than 4 or 5 apps open when building servers, this suits my style.

Just like the shutdown tracker, this setting is stored in the registry. A one liner for this would look like this:

Set-ItemProperty -Path 'HKCU:\Software\Microsoft\Windows\CurrentVersion\Explorer\Advanced' -Name TaskbarGlomLevel -Value 0

0 is the value for “always combine”, 1 for “combine when taskbar full” and 2 for “never combine”. In order for the setting to take effect, one of two things has to happen. Either log off/restart, or restart the explorer.exe process. The later can be performed by running the following:

Stop-Process -ProcessName explorer -force

If you’d like to use a function for this, we can use something like the code below in our server build script:

function Set-TaskbarGrouping {
    [CmdletBinding(SupportsShouldProcess, SupportsPaging, DefaultParameterSetName = 'NeverCombine')]
    param(
        # Always combines similar shortcuts into groups
        [Parameter(ValueFromPipelineByPropertyName, ParameterSetName = 'AlwaysCombine')]        
        [switch] $AlwaysCombine,
        
        # Combines similar shortcuts into groups only when the taskbar is full
        [Parameter(ValueFromPipelineByPropertyName, ParameterSetName = 'CombineWhenTaskbarFull')]
        [switch] $CombineWhenTaskbarFull,
        
        # Never combines similar shortcuts into groups
        [Parameter(ValueFromPipelineByPropertyName, ParameterSetName = 'NeverCombine')]
        [switch] $NeverCombine,
        
        # Restarts explorer in order for the grouping setting to immediately take effect. If not specified, the change will take effect after the computer is restarted
        [switch] $NoReboot
    )
    switch ($PsCmdlet.ParameterSetName) {
        'AlwaysCombine' {
            Set-ItemProperty -Path 'HKCU:\Software\Microsoft\Windows\CurrentVersion\Explorer\Advanced' -Name TaskbarGlomLevel -Value 0
        }
        'CombineWhenTaskbarFull' {
            Set-ItemProperty -Path 'HKCU:\Software\Microsoft\Windows\CurrentVersion\Explorer\Advanced' -Name TaskbarGlomLevel -Value 1
        }
        'NeverCombine' {
            Set-ItemProperty -Path 'HKCU:\Software\Microsoft\Windows\CurrentVersion\Explorer\Advanced' -Name TaskbarGlomLevel -Value 2
        }
    }
    if ($NoReboot){
        Stop-Process -Name explorer -Force
    }else{
        Write-Verbose -Message 'Change will take effect after the computer is restarted'
    }
} # end function Set-TaskbarGrouping

I use parameter set names so that only one of the parameters can be used when the function is called. The three options are “NeverCombine” “CombineWhenTaskbarFull” and “AlwaysCombine”. But since I define the parameters in a param block, you get tab completion. So no need to even remember the options. For example:

Set-TaskbarGrouping -NeverCombine

If you also include the -NoReboot parameter when calling the function, it will restart explorer.exe to avoid the need to log off/restart.

When you spend time building servers, there are often some minor tweaks that you use to make life easier. In many environments, Group Policy Objects (GPOs) are used to configure these settings. But in a lot of environments, that’s not the case. If you build a lot of servers, you may have some scripts to help streamline the process. I often see this being the case among consultants who are engaged to deploy a solution. If you’ve followed my blog for a while, you know that’s what I do. And I look for many ways to streamline the deployment. Many solutions I write are all about the actual deployment, whereas this particular post is about the working environment I’ll be spending time in.

One thing that always drives me nuts is the Shutdown Tracker. That’s the little dialog box that pops up when you want to restart or shutdown a server. You’re presented with a prompt to pick the reason why you’re restarting or shutting down. While this can certainly have its place in an enterprise environment, it’s not generally needed during a deployment. And it’s not likely needed in a lab environment where you might be testing various configurations and restarting often. So let’s gag that annoying prompt.

To disable Shutdown Tracker, open an elevated PowerShell prompt and enter the following one line:

Set-ItemProperty -Path "HKLM:\SOFTWARE\Policies\Microsoft\Windows NT\Reliability" -Name ShutdownReasonOn -Value 0

This will take care of the problem. If you later want to enable the Shutdown Tracker, you can simply run it again, specifying a 1 for the value.

We can make this a little more flexible by creating a function to let us enable or disable as needed.

function Set-ShutdownTracker {
	[CmdletBinding(SupportsShouldProcess = $True, SupportsPaging = $True, DefaultParameterSetName = "disabled")]
	param(
		# Disable the shutdown tracker
		[Parameter(ValueFromPipeline = $False, ValueFromPipelineByPropertyName = $True, ParameterSetName = "disabled")]
		[switch] $Disabled,
		
		# Enable the shutdown tracker
		[Parameter(ValueFromPipeline = $False, ValueFromPipelineByPropertyName = $True, ParameterSetName = "enabled")]
		[switch] $Enabled
	)
	switch ($PsCmdlet.ParameterSetName) {
		"enabled" {
			Set-ItemProperty -Path "HKLM:\SOFTWARE\Policies\Microsoft\Windows NT\Reliability" -Name ShutdownReasonOn -Value 1
		}
		"disabled" {
			Set-ItemProperty -Path "HKLM:\SOFTWARE\Policies\Microsoft\Windows NT\Reliability" -Name ShutdownReasonOn -Value 0
		}
	}
} # end function Set-ShutdownTracker

And the script can be called with either the -Enabled or -Disabled parameters.

Adding the one liners or the function to your deployment scripts might make life a little easier.

Description

There are some really helpful scripts out there. Not just for Lync and Exchange. But many other apps and administrative tasks. Sometimes, however, the people who need to run them aren’t well versed in PowerShell. This makes is cumbersome for them to open PowerShell, navigate to a folder containing a script, and execute it with the correct parameters. This often leads to complaints about the difficulty of the process, or those admins just not using that tool. As not all admins have our PowerShell prowess, we can create a desktop shortcut that will allow an admin to simply double-click on it to execute it. Let’s see an example.

For this example, I’m going to use Johan Veldhuis’ very slick sefautil GUI, a wrapper for the Lync sefautil.exe program. Sefautil is a resource kit utility that allows an admin to set things like delegates, call forwarding, and other settings, on behalf of users. Sefautil has some really painful syntax, and a complete lack of error reporting. Using it is often frustrating. Johan’s GUI for it makes life SO much easier, that I found myself using it a LOT.

Let’s say, for the sake of this example, that the script, called sefautil_gui.ps1, is in a folder called c:\_scripts. When you execute Johan’s script, you must pass it a front end pool name using the “-pool” parameter. Normally running it would require something like the following:

.\sefautil_gui.ps1 -pool pool01.contoso.com

With a shortcut, we need to tell it to launch PowerShell, and call the script along with the parameters. The syntax is the full path to powershell.exe, along with the “-command” parameter and the syntax used for the script. The syntax is wrapped in quotes, and prefixed with an ampersand. PowerShell resides at C:\Windows\System32\WindowsPowerShell\v1.0\powershell.exe

1. Right click on the desktop, and choose New>Shortcut.
2. Enter a path for the shortcut. For our example, we’ll use
   C:\Windows\System32\WindowsPowerShell\v1.0\powershell.exe -command “& c:\_scripts\sefautil_gui.ps1 -pool pool01.contoso.com”
3. Click Next.
4. Give the shortcut a name. We’ll call it “sefautil GUI”. Then click Finish.

Let’s set the starting path. Right click on the newly created shortcut, and click Properties. Click on the Shortcut tab. In the “Start In” field, let’s set it to “C:\windows\system32\WindowsPowerShell\v1.0”.

You may have noticed that the shortcut has the PowerShell icon. While that’s all fine and damn sexy, we might want to change that. That’s simple enough since we’re already on the Shortcut tab, click the Change Icon at the bottom and choose whichever icon you’d like. For mine, I chose the Deployment Wizard icon. This is available by browsing to %ProgramFiles%\Microsoft Lync Server 2013\Deployment and choosing Bootstrapper.exe.

While Johan’s script is a GUI based script, many are not. If that’s the case, we can tweak the session settings a little further. On the Options tab of the shortcut, you can tailor settings like Quick Edit mode, which makes selecting, copying, and pasting easier. Obviously, the Font, Layout, and other tabs provide further control over the experience.

Also note that non-GUI scripts will close the PowerShell window when they are done running, so a script might need to be tweaked to pause before closing. YMMV.

Once you’re done, click OK. Voila!

Now, simply double clicking on our new shortcut launches the script.

I often do this for many 3rd party administrative tools, including Lync Call Pickup Group Manager, Lync RGS Holiday Set Editor, Centralized Logging Tool, and more.

Donations

I’ve never been one to really solicit donations for my work. My offerings are created because *I* need to solve a problem, and once I do, it makes sense to offer the results of my work to the public. I mean, let’s face it: I can’t be the only one with that particular issue, right? Quite often, to my surprise, I’m asked why I don’t have a “donate” button so people can donate a few bucks. I’ve never really put much thought into it. But those inquiries are coming more often now, so I’m yielding to them. If you’d like to donate, you can send a few bucks via PayPal at https://www.paypal.me/PatRichard. Money collected from that will go to the costs of my website (hosting and domain names), as well as to my home lab.