https://technet.microsoft.com/en-us/library/dn280950.aspx

   This page cannot be found anywhere on the file system. It’s dynamically generated (based on your identity providers) by ADFS at run time. Because of this, you can’t find the page on the file system and modify it.

    Now, to be fair, Microsoft made their HRD page super easy to customize as defined in this TechNet article, titled “Customizing the AD FS Sign-in Pages”.

     You can change the logo image, illustration image, description text, etc. Unfortunately, that’s not good enough for this project. We really want to make this page integrate with the feel of the overall application. Just changing logo and description enough. Luckily, Microsoft gives us the tools we need, to really do anything we want on this page.

   Before I get started, I want to say that I don’t want a “hack solution”. I won’t destroy or lose any logic that the page uses already. I just want to tailor that logic into my “custom” page.

    In addition to being able to set the basic properties described in the TechNet article listed above, we can also set JavaScript to be called in the page. The JavaScript is not in a function, it just drops it in a script tag in the body. This means that it will run when the page loads. You will need some basic JavaScript experience, but I’ll do most of the heavy lifting for you.

To set the JavaScript on this page use the PowerShell cmdlet:

Set-AdfsWebTheme (click for documentation)

We will use the –AdditionalFileResource flag to specify we want to load some JavaScript in the page.

 Here is a sample:

Set-AdfsWebTheme -TargetName ADAM -AdditionalFileResource @{Uri=’/adfs/portal/script/onload.js’;path='C:\ZnHub PowerShell\DefaultWebTheme\script\onload.js'}

This will load my code JavaScript code located in “C:\ZnHub PowerShell\DefaultWebTheme\script\onload.js”. Ok.. great! so what.. well, we can do everything from here! Let’s take a look at some of my code.

First I want to hide (but not destroy the existing code!):

var cw = document.getElementById('contentWrapper').style.diplay = "none";

var bw = document.getElementById('brandingWrapper').style.display = "none";

var fp = document.getElementById('fullPage').style.height = 0;

I’m finding the elements that they use, and hiding them. I set the fulPage l <div> to a height of 0, because I don’t’ want any unnecessary scrollbars to pop up.

Now I want to create my page. I am going to do this by creating a new <div> tag. I’ll put all my HTML page markup code in that tag. Finally, I will add that <div> to the body element of the JavaScript DOM.

Here’s a look at some code:

var iDiv =  "<nav class='navbar navbar-default'><span><a  class='navbar-brand' href='https://www.microsoft.com'><img style='margin-top:-4px' title='Microsoft' src='/adfs/portal/logo/logo.png' alt='' /></a><a class='navbar-brand'>ZN Sample Services</a><span></nav></br></br></br><center><a class='btn btn-success btn-lg' onclick=\"HRD.selection('https://corp.sts.zn.com')\">Zn Employee Sign In</a>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;<a class='btn btn-info btn-lg' onclick=\"HRD.selection('https://vmastest.accesscontrol.windows.net/')\">Sign in with Windows Live</a></br></br></br><img class='largeIcon float' src='/adfs/portal/illustration/illustration.png' height=500/></center>";

Believe it or not, that’s my whole page! And it’s beautiful! Now some notes:

First of all, the code has to be on one line. You cannot have carriage returns when building your <div> like I do.

Second, no double quotes are allowed.

 So

<a href = ”test” >

Becomes

<a href = ‘test’ >

Now there is one super catch this. When calling a function, you have use double quotes so you must escape them like this:

<a onclick=”test()”>

Becomes

<a onclick=/”test()/”>

Be careful when working on the page, do small parts and test often!

One thing you’ll notice in my HTML code, is that I am using a lot of styles. Better believe it!. there is a cmdlet that allows you to specify the CSS file you want the HRD to use:

SetAdfsWebTheme –StyleSheet

Use it just like we did to load the JavaScript. Now you have use all the styles that you use in your app, in the HRD! Very powerful.

My code above is very small. That’s because most of the content is static.. it doesn’t ever change. Instead of trying to HTML mark up all this static conent.. I just took a screenshot capture of the static content. I then use that image to display all the content. This is much easier than building all that HTML ark up in the <div> tag (though you can).

To use an image like I did, you can use the SetADFSWebTheme –Illistration cmdlet. Same syntax and use. Now you can use that image with the following source:

src='/adfs/portal/illustration/illustration.png'

Really simple… don’t forget you can also use the SetADFSWebTheme –Logo cmdlet to load images as well.

An even better solution, especially if you don't want to touch the default theme is to use, and if you have to add lots of buttons or style sheets, is to use:

Set-AdfsWebTheme-TargetNameVS-AdditionalFileResource @{Uri=’/adfs/portal/images/live.png';path='C:\HRD\live.png'} 

This is really they key to adding resources. By specifying a Uri (the cmdlet function takes an array of hashtables). Now we can access the resources in our code like this:

<img source=adfs/portal/images/live.png'/>

This cmdlet takes and file type, so you can do this with anything you need to build your page.

The last part of my code is wiring up our buttons just the default HRD so they send me off to login. Remember earlier, I said that I don’t want to do anything funky, I just want to do the same thing the original page did. So let’s look at the source of the default HDR page. Here we see that what happens, when a provider is selected, we call the HRD.selection JavaScript function with the correct realm, which in turn then posts the form back to ADFS which the correct payload, and ADFS takes over from there.

Remember we didn’t clobber any original page code, so we can do EXACTLY as the original page does and call that JavaScript function:

HRD.selection('https://zntest.accesscontrol.windows.net/')

This will do exactly the same thing as their buttons do when clicked (essentially call the JavaScript HRD function and post the form back to ADFS.

To find your realms, bring the default HRD page up in a browser. In the options, view the source of the page and search for HRD.selection. This will yield you your realms as they should be used.

If you wanted the HRD page to be totally skipped, and go directly to the zntest provider for exmaple, you just put this code in the JavaScript:

HRD.selection('https://zntest.accesscontrol.windows.net");

And you’re done!.. as soon as the page loads, it will go to the zntest realm.

I use it in a button click like this:

<a onclick=\"HRD.selection('https://zntest.accesscontrol.windows.net/')\">

The last thing we have to do is add our newly created <div> tag to the body element:

document.body.innerHTML += iDiv; 

Here is a look at my finished page, note I had to cut up actual images and text, but you get it:

As always hope this helps!

There comes a time when we will need to have buttons (or radioButtons, any control) that needs to have it's width dynamically assigned based on the content within, for example a button with the text "MyButton, should be wide enough to display all the text, but another button next to it should have the same padding but a smaller width if it's content was "Small". This gets even more confusing when we have a list of items that are data bound. You cannot set individual widths for items that are not yet created.

I'm going to show you how you can accomplish this using a MVVM Converter (https://msdn.microsoft.com/en-us/library/system.windows.data.binding.converter(v=vs.110).aspx).

First let's define our list: 

public ObservableCollection<string> ButtonNames       
{
     get { return _buttonNames; }
}

private ObservableCollection<string> _buttonNames = new ObservableCollection<string>();

Let's add some strings to our ButtonNames dependency property:

 _buttonNames.Add("Short");
 _buttonNames.Add("LongerName");
OnPropertyChanged("ButtonNames");

Pretty simple.. now we'll create an ItemsControl to bind to our ButtonNames collection:

<ItemsControl ItemsSource="{Binding ButtonNames}" >
   <ItemsControl.ItemsPanel>
      <ItemsPanelTemplate>
         <StackPanel Orientation="Horizontal"/>
      </ItemsPanelTemplate>
   </ItemsControl.ItemsPanel>
   <ItemsControl.ItemTemplate>
      <DataTemplate>
         <Button Content="{Binding}" Width="50" Height="30" />
      </DataTemplate>
   </ItemsControl.ItemTemplate>
</ItemsControl>

If we run this, we see our problem:

The buttons are stuck with the wrong width.(say that fast three times!)

To fix this, I am going to bind the width of our button to the string we are using for content. The problem is, the property is expecting an int, So we need a converter:

class WidthHelper : IValueConverter
{
     public object Convert(object value, Type targetType, object parameter, System.Globalization.CultureInfo culture)
          {
                int nameLength = value.ToString().Length;
                return (14 * nameLength) + 20;
          }

     public object ConvertBack(object value, Type targetType, object parameter, System.Globalization.CultureInfo culture)
          {
                throw new NotImplementedException();
          }
}

What's great about the implemented Convert function is that the value coming in is an object, so we can pass in a string.. or whatever we want. As long as we match the targetType.. we're good. So as you can see I multiply 14 (that's the font size I am using) times the number of characters in the name, I then put a 10 pixel buffer on each side of the text (+20).

I can now wire up ourconverter to our ItemsControl data template:

<Button Content="{Binding}" Width="{Binding Converter,Converter={StaticResource WidthHelper}}" Height="30" />

Our buttons now display with the dynamic width:

Pretty cool!

 Hope this helps!

  I have a user control in Silverlight, I'm using C# on the back end. The control's background color changes via viewmodel databinding. What I want to do, is get a callback when the background color changes. so I can start my storyboard which is a nice transition into the UI change of the control.

  Here's an idea of my control:

<Grid x:Name="LayoutRoot" Background="{Binding Background}" Height="85" Width="300" >

        <Grid.Resources>

            <Storyboard x:Name="animate">

                <DoubleAnimationUsingKeyFrames Storyboard.TargetProperty="(UIElement.Opacity)" Storyboard.TargetName="LayoutRoot">

                    <EasingDoubleKeyFrame KeyTime="0" Value="0"/>

                    <EasingDoubleKeyFrame KeyTime="0:0:1.0" Value="1"/>

                </DoubleAnimationUsingKeyFrames>

            </Storyboard>

        </Grid.Resources>

    ...

   </Grid>

   I spent a lot of time trying to do things like hook into the BIndingExpression and trying to wire up different ways to hook up to this control. Nothing, was, for lack of better terms, code pretty or really functional. I also really only care about when the background property on the Grid control changes. So how can this be accomplished? The answer is that you don't need to look up anything, you have all the knowledge to do this, you just have to think about it.

   The Grid control is a bad control for what we want.. it just doesn't give me what I need, a callback when the data (that again is Databound somewhere in a viewmodel) changes. But think about it.. other controls do have this functionality! The textbox control for one does! And that works perfectly for us. Let's add a textbox control to our user control:

<TextBox TextChanged="TextBox_TextChanged" Text="{Binding Background}" Opacity="0"/>

   The textbox control will now callback into the code anytime the Background (which in my case is a Solid Color brush) changes. Now all I have to do is wire up the event handler:

private void TextBox_TextChanged(object sender, TextChangedEventArgs e)

{

     animate.Begin();

 }

    So pretty easy to do, and the code is not freaky.. it's straight forward and consistent. It also allows me to wire up many textbox controls if necessary to monitor different property changes.

  Enjoy!

I have found samples on the internet for each, but not a comprehensive example for both. Hopefully putting these pieces all together for you will save you some time!

First we must declare the dependency property. It must be an attached dependency property to be used in xaml. In this sample I am going to create a button that has a new property called HasPlayButton and another property called HasStopButton to display or hide a path on a checkbox control. This sample will work for Silverlight, Windows Store Apps, Windows Phone Apps and WPF apps, pretty much anything that uses XAML.

First we'll make our new Checkbox control. This is just a simple control that inherits from the Checkbox control. I will be adding two dependency properties to this class:

partial class AppButtton : Checkbox
{
  public static readonly DependencyProperty HasPlayButtonProperty = DependencyProperty.RegisterAttached("HasPlayButton", typeof(Visibility), typeof(AdamButton), new PropertyMetadata(Visibility.Visible));

    public static void SetHasPlayButton(AdamButton element, Visibility value)  
    {

            element.SetValue(HasPlayButtonProperty, value);
    }

  public static Visibility GetHasPlayButton(AdamButton element)
  {
         return (Visibility)element.GetValue(HasPlayButtonProperty);
  }

    publicVisibility HasPlayButton
  {
         get { return (Visibility)GetValue(HasPlayButtonProperty); }

    set { SetValue(HasPlayButtonProperty, value); }
  }

  public static readonly DependencyProperty HasStopButtonProperty = DependencyProperty.RegisterAttached("HasStopButton", typeof(Visibility), typeof(AdamButton), new  PropertyMetadata(Visibility.Collapsed));

  public static void SetHasStopButton(CheckBox element, Visibility value)
    {
        element.SetValue(HasStopButtonProperty, value);
    }

  public static Visibility GetHasStopButton(CheckBox element)
    {
       return (Visibility)element.GetValue(HasStopButtonProperty);
    }

  public Visibility HasStopButton
    {
        get { return (Visibility)GetValue(HasStopButtonProperty); }

    set { SetValue(HasStopButtonProperty, value); }
    }

  public AdamButton()
    {
         InitializeComponent();
    }
}

 Let's look at the definition of the first dependency property:

   public static readonly DependencyProperty HasPlayButtonProperty = DependencyProperty.RegisterAttached("HasPlayButton", typeof(Visibility), typeof(AdamButton), new PropertyMetadata(Visibility.Visible));

You can see that I have defined this as an attached property:

    DependencyProperty.RegisterAttached

Next I have to set the type for the value that this property will use. I am using the Visibility type so I can hide my path elements later.

The next parameter is the type that I am setting this property on, which is AdamButton (my custom control).

The last parameter is the default value of this property, which we pass in via the PropertyMetadata class.

new  PropertyMetadata(Visibility.Collapsed


Since this is an attached dependency property, we must implement both the Setter and Getter for the property, note, you must use the naming convention followed below: 

  public static void SetHasPlayButton(AdamButton element, Visibility value)  
    {

            element.SetValue(HasPlayButtonProperty, value);
    }

  public static Visibility GetHasPlayButton(AdamButton element)
  {
         return (Visibility)element.GetValue(HasPlayButtonProperty);
  }


The last step to complete our custom button is to define the regular normal property on the AdamButton class. This allows our class to be aware of these properties as well.

  publicVisibility HasPlayButton
  {
         get { return (Visibility)GetValue(HasPlayButtonProperty); }

    set { SetValue(HasPlayButtonProperty, value); }
  }

Our class is now complete and we may use the AdamButton class with the new HasStopButton and HasPlayButton properties any where in xaml.

I am going to create a style for the AdamButton control. This style will implement the new properties we defined above via Template binding in xaml. Before we can use our AdamButton class in xaml, we must first declare it's namespace:

xmlns:a="clr-namespace:PhoneApp2"

Add this to your root node for your application or app.xaml file.


Now we can user our new properties via the AdamButton class in a style. Let's take a look at the Style:

<Style x:Key ="PlayButtonStyleLarge" TargetType="CheckBox">
            <Setter Property="Background" Value="#FF448DCA"/>
            <Setter Property="BorderThickness" Value="0"/>
            <Setter Property="MinWidth" Value="50"/>
            <Setter Property="MinHeight" Value="19"/>
            <Setter Property="Template">
                <Setter.Value>
                    <ControlTemplate TargetType="a:AdamButton">
                        <Border BorderBrush="{TemplateBinding BorderBrush}" BorderThickness="{TemplateBinding BorderThickness}" >
                            <VisualStateManager.VisualStateGroups>
                                <VisualStateGroup x:Name="CommonStates">
                                    <VisualState x:Name="Normal"/>
                                    <VisualState x:Name="MouseOver" />
                                    <VisualState x:Name="Pressed" />
                                    <VisualState x:Name="Disabled" />
                                </VisualStateGroup>
                                <VisualStateGroup x:Name="CheckStates" />
                                <VisualStateGroup x:Name="FocusStates">
                                    <VisualState x:Name="Focused" />
                                    <VisualState x:Name="Unfocused" />
                                </VisualStateGroup>
                            </VisualStateManager.VisualStateGroups>
                            <Grid>
                                <Border x:Name="ButtonBackground" BorderBrush="{TemplateBinding BorderBrush}" BorderThickness="{TemplateBinding BorderThickness}" Background="{TemplateBinding Background}" CornerRadius="0" Margin="{StaticResource PhoneTouchTargetOverhang}">
                                    <Grid>
                                        <Polygon Points="0,00 50,30, 0,65" Fill="White" Name="StartPoly" VerticalAlignment="Center" HorizontalAlignment="Center" Visibility="{TemplateBinding HasPlayButton}" />
                                        <Polygon Points="0,0 50,0, 50,50 0,50" Fill="White" Name="StopPoly"  VerticalAlignment="Center" HorizontalAlignment="Center" Visibility="{TemplateBinding HasStopButton}"/>
                                        <ContentControl x:Name="ContentContainer" ContentTemplate="{TemplateBinding ContentTemplate}" Foreground="{TemplateBinding Foreground}" HorizontalContentAlignment="{TemplateBinding HorizontalContentAlignment}" Padding="{TemplateBinding Padding}" VerticalContentAlignment="{TemplateBinding VerticalContentAlignment}" />
                                    </Grid>
                                </Border>
                            </Grid>
                        </Border>
                    </ControlTemplate>
                </Setter.Value>
            </Setter>
        </Style>      

Notice the target type for the style is still set to Checkbox, because that is our base class style that xaml (and .net) can recognize:

<Style x:Key ="PlayButtonStyleLarge" TargetType="CheckBox"> 

However the control template is set to our AdamButton class. This allows me to access my new properties via the control's template binding:

<ControlTemplate TargetType="a:AdamButton">

The "a:" here is our namespace reference we defined earlier.

In my control style I set the visibility of the different paths with our new property:

<Polygon Points="0,00 50,30, 0,65" Fill="White" Name="StartPoly" VerticalAlignment="Center" HorizontalAlignment="Center" Visibility="{TemplateBinding HasPlayButton}" />



We now have a button that we can easily style with our new properties in C# code behind. Here's an example of how we can use the AdamButton class:

if (<value>== true)
{
        Button1.HasPlayButton = Visibility.Collapsed;
    Button1.HasStopButton = Visibility.Visible;
}

else
{
       Button1.HasPlayButton = Visibility.Visible;
       Button1.HasStopButton = Visibility.Collapsed;
}

Hope this helps guys!

A

SCVMM 2012 has replaced it's pssnapin (formerly Microsoft.SystemCenter.VirtualMachineManager) with a PowerShell module. Modules were introduced in PowerShell v2 and have added functionality over pssnapins. There is plenty on the web on PowerShell modules, and that is not the topic of this blog. In this blog I will show you how to load the SCVMM 2012 PowerShell module so you may make C# calls to SCVMM 2012 via embedded PowerShell.

In my applications, I use the InvokePowerShell function to call into SCVMM any time I need to work with any of the objects managed by SCVMM. My function will take a string parameter that contains the powershell I want to call. In my sample string you will see I add \r\n in my strings. This puts line breaks in the string so that when I debug, and I view my powershell in the Visual Studio debugger windows, the script is very easy to read (and check for errors!).

The function will return a PSObject. This PSObject will return any errors if occurred, or if successful any information that would be normally seen in the powershell window. I show some examples of how to use this function as well.

Before you can use this function, you will need to add a reference to the System.Management.Automation.dll. This is not so straight forwarded because this .dll is a little hard to find. Take a look at this article that will help you add this reference.

https://www.codeplex.com/wikipage?ProjectName=ShinyPower&title=Frequently%20Asked%20Questions

Here is link for reference on the System.Management.Automation namespace:

https://msdn.microsoft.com/en-us/library/system.management.automation(VS.85).aspx

OK.. let's take a look at the function:

RunspaceConfiguration rconfig = RunspaceConfiguration.Create();
 PSSnapInException Pwarn = new PSSnapInException();
 Collection<PSObject> results = new Collection<PSObject>();
 Runspace runspace = RunspaceFactory.CreateRunspace();
 
 string test = "Import-Module VirtualMachineManager\r\n";
 runspace = RunspaceFactory.CreateRunspace(rconfig);
 runspace.Open();
 
 pipeline = runspace.CreatePipeline();
 pipeline.Commands.AddScript(test);
 
 
 try
 {
 results = pipeline.Invoke();
 }
 catch (Exception ex)
 {
 return ex.Message;
 }

OK so now we have a function to call our powershell scripts. You will have to check and handle your exceptions yourself. Do not leave the try-catch like this as you will miss uhandled exceptions. 

There is one issue you have to be very careful with with this function. If this function is run in a process, and multiple calls to Get-VMMSerrver are called, a connection for each call will be held in the process until the process is taken down. The runspace Close and Dispose call do not clear these out. Avoid calling the generic Get-VMMServer cmdlet with this function. If you must use this cmdlet, make sure that it is only called once, and store a way to check with it with a static variable in your code.

Let's take a look at using the InvokePoweShell function. I'll use this function to take a SCVMM host out of placement. I will pass into this function a string for the host name, and a boolean value to set the host available and not available.  The function will return true, if it succeeds and false if there is a problem:

public bool SetHostAvailabeForPlacement(string HostName, boolean val)

        {

            bool retval = true;

            string psscript = "$MyHost = Get-VMHost -VMMServer localhost -ComputerName '" + HostName + "'\r\n Set-VMHost -VMHost $MyHost -AvailableForPlacement $" + val;

            try

            {

                Collection<PSObject> results = InvokePowerShell(psscript);

            }

            catch

            {

                return false;

            }

            return retval;

        }

You will notice that in my powershell script, I didn't use the Get-VMMServer function instead I just told the cmdlet which server to use. I also could have made this powershell a little cleaner by making the cmdlet call in one line:

string psscript = "$MyHost = Get-VMHost -VMMServer localhost -ComputerName '" + HostName + "' | Set-VMHost -AvailableForPlacement $" + val;

In this example I'm not looking for any data back so I don't check the results PSObject. See my other blog posts for examples of this.

The last thing I want to demonstrate in this blog is how to use powershell that can be called from C# where you don't need to recompile your code to change the powershell. What I do is read the contents of an external text file into a string. I then parse that string replacing place holders with the values that I need.

public void CallPSFromFile(string FileName)

{

            StreamReader sr = new StreamReader(FileName);

            string psscript = sr.ReadToEnd();

            sr.Close();

            string VMName = "MyTestVM";

            psscript.Replace("__VMName__", VMName);

            Collection<PSObject> results = InvokePowerShell(psscript);

}

This is a very simple function, and in the real world, you will want to make sure that your file read process doesn't error out.  The StreamReader is the System.IO namespace, add a Using statement to add it to your code.

This function will read in the text from the text file and then replace any occurances of "__VMName__" with my value. I add the underscores to my replacement text to make them easy to find.

Here's a sample text file to remove a VM that can be used with the above sample:

$MyVMs = @(Get-VM -VMMServer VMMServer1.Contoso.com | where { $_.Name -Match "__VMName__" } )

$MyVMs | Remove-VM

As always, enjoy!

   This post was written using .NET Framework 4.0 and storing my data in SQL Server 2008.

   Let's take a look at the problem I had at hand. I needed to be able to store Key\Value pairs from a C# application into SQL. Since I did not know how many key value pairs I might end up with I decided to use the Dictionary object in c#. The problem is, both the Dictionary and Hashtable classes or not serializabe because they implement the IDictionary interface. Here's the error that you will get if you try to serialize one of the classes:

System.NotSupportedException was unhandled

  Message=The type System.Collections.Generic.Dictionary`2[[System.String, mscorlib, Version=2.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089],[System.String, mscorlib, Version=2.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089]] is not supported because it implements IDictionary.

  Source=System.Xml

  StackTrace:
       at System.Xml.Serialization.TypeScope.GetDefaultIndexer(Type type, String memberInfo)
       at System.Xml.Serialization.TypeScope.ImportTypeDesc(Type type, MemberInfo memberInfo, Boolean directReference)
       at System.Xml.Serialization.TypeScope.GetTypeDesc(Type type, MemberInfo source, Boolean directReference, Boolean throwOnError)
      ...

   The reason the System.NotSupportedException is thrown is because the IDictionary class implements the IXmlSerializable interface.The IXmlSerializable interface makes us overide the following methods: GetSchema, ReadXml and WriteXml. We can do this, but since we have to write a class to overide these functions anyway, it's easier to just write a generic class and use that for serialization. Let's take a look.

   What we are going to do is make a stub class that will be serializbe that mimics the Dictionary or Hashtable DictionaryItem or KeyValuePair<string, string>. We want this class to be a simple a possible:

public class DataItem
{

   public string Key;

   public string Value;

   public DataItem(string key, string value)

   {

      Key = key;

      Value = value;

   }

List<DataItem> tempdataitems = new List<DataItem>(myDictionary.Count);

   Since this is now serializable, we can jjust foreach through our Dictionary or Hashtable objects and build our tempdataitmes class to serialize.

foreach (string key in myDictionary.Keys)

{

    tempdataitems.Add(new DataItem(key, myDictionary[key].ToString()));

}

private string SerializeJobData()

{

   List<DataItem> tempdataitems = new List<DataItem>(myDictionary.Count);

   foreach (string key in dataitems.Keys)

   {

       tempdataitems.Add(new DataItem(key, dataitems[key].ToString()));

   }

   XmlSerializer serializer = new XmlSerializer(typeof(List<DataItem>));

   StringWriter sw = new StringWriter();

   XmlSerializerNamespaces ns = new XmlSerializerNamespaces();

   ns.Add("","");

   serializer.Serialize(sw, tempdataitems,ns);

   return sw.ToString();

}

   In this example I am using the .net 4 StringWriter class from the System.IO namespace. This makes working with the stream needed for xml serialization simple because it hanldes the stream needed for xml serialization. In addtion you can see that I didn't want the default .net namespace in my xml so I set the namespace of the xmlsericalizer  to "". If you want more control over your xml be sure to include your custom namespace.

   Here's what the data looks like in SQL:

<ArrayOfDataItem>

  <DataItem>

    <Key>Vera</Key>

    <Value>Vera</Value>

  </DataItem>

  <DataItem>

    <Key>David</Key>

    <Value>David</Value>

  </DataItem>

  <DataItem>

    <Key>Susan</Key>

    <Value>Susan</Value>

  </DataItem>

  <DataItem>

    <Key>FakeParam</Key>

    <Value>FakeParam</Value>

  </DataItem>

</ArrayOfDataItem> 

   As you can see this is nice and clean.

   The last thing we would need to tacke is deserializing back to a Dictionary<string, string> object. I'm going to again use the StringWriter class to help me.

private void DeserializeData(string RawData)

{

   Dictionary<string, string> myDictionary = new Dictionary<string, string>();

   XmlSerializer xs = new XmlSerializer(typeof(List<DataItem>));

   StringReader sr = new StringReader(RawData);

   List<DataItem> templist = (List<DataItem>)xs.Deserialize(sr);

   foreach (DataItem di in templist)

   {

       myDictionary.Add(di.Key, di.Value);

   }

}

   Super simple!

   I hope this helps you quicly get through this task and answers all questions regarding serializing and deserialize the .net 4 Dictionary and Hashtable classes.

   In this post I will discuss a question that has come my way a few times. This post will discuss how to get the percentage of a job task from SCVMM that you may use in your custom applications to allow your end users to watch the progress of a task performed on a VM being managed by SCVMM. This post will assume you are familiar with programming against SCVMM via both powershell and C#. If you are not familiar with these concepts, please review my previous blog posts to get started.

   I will not delve into the understanding of the SCVMM tables stored in SQL. I wll highlight which tables contain the information we need to perform our desired task. Please note, that any modifcation to the SCVMM database (default name VirtualManagerDB) is unsupported. Any modifications to this database may break your existing SCVMM setup, and Microsoft may not assist you with recovery of this data, as modifications to this database are unsupported.

   The first table we will need to look is the "tbl_WLC_VObject table". This tables contains base information for SCVMM objects (VMs, Templates, etc). You can differentiate the different objects by their object type found in this table. For example, objecttype "1" is a virtual machine, object type "3' is a template. The field we're going after is the "FailedJobID" field. This field will either contain a NULL value or a GUID. We will need this GUID to look up the progress status. There are tons of different ways you may wish to filter for this GUID, but to keep it simple, I'll just go with:

SELECT [FailedJobID]

  FROM [VirtualManagerDB].[dbo].[tbl_WLC_VObject]

  WHERE ObjectType = 1 AND Name = 'MyGreatVM'

  
   So far so good. We now have a GUID to represent the JobID for the job in question. From this point on I will refer to this GUID as JobID.

   The last thing we will need to do is query the table called "tbl_TR_SubtaskTrail" to get our completion status. Remember that this will continuously change as the status progresses so you will want to keep comming back and checking this table until your task is done. Here's what we will use:

    I've added a bit more logic in my query to insure that I am targetting a task description of "NewVM". In this example, we will assume that we are going to get progress status for a new VM that is building. The addition of the Description ='NewVM'  feild insures that I only get progress for this particular task (the New VM build task).

SELECT PercentComplete

FROM tbl_TR_SubtaskTrail

WHERE tbl_TR_SubtaskTrail.TaskID = @JobID

AND Description = 'NewVM'

  
   That's it, you can now use this display data wherever you like to keep you clients informed of the percentage complete of a task in SCVMM. I have also showed how to give a little extra insurance that you are tracking the task that you are looking for by using an extra WHERE clause against the Description field.

   In reality you will probably not want to call this SQL statement from code in a loop unitl you hit 100 percent. You will want to instead create a stored procedure instead in SQL and call that stored procedure with a parameter value of JobID. This is the least expensive way to get the data, and allows SQL to do it's job (which is all the work).

   As always enjoy!

https://blogs.msdn.com/taylorb/archive/2008/07/29/hyper-v-wmi-creating-a-thumbnail-image.aspx

    Let's talk about a few things with my blog post. First you can call into SCVMM to get thumbnails.

    We could call into SCVMM and get a thumbnail in powershell but I find that the performance is not nearly as fast as calling directly into Hyper-V. Performance for me is paramount because I am going to be refreshing my thumbnails often. I also prefer not to have to queue many SCVMM cmdlets, I need SCVMM available and ready to service my VMs for actions such as Stop-VM or Remove-VM. Instead I am going to call directly into Hyper-V via WMI calls. Use WMI calls with caution, and you should be comfortable and have a working understanding of WMI calls to fully understand this blog post.

    The function I am going to write, GetThumbnailForVM takes two parameters:

    vmName - string which is the name of the VM we want to get the thumbnail for.

    vmHost - string which tells us what host the VM lives on

    You could dynamically get the vmHost by calling:

Get-VM [vmName]| select VMHost

In my applications, I use the InvokePowerShell function to call into SCVMM any time I need to work with any of the objects managed by SCVMM. My function will take a string parameter that contains the powershell I want to call. In my sample string you will see I add \r\n in my strings. This puts line breaks in the string so that when I debug, and I view my powershell in the Visual Studio debugger windows, the script is very easy to read (and check for errors!).

The function will return a PSObject. This PSObject will return any errors if occurred, or if successful any information that would be normally seen in the powershell window. I show some examples of how to use this function as well.

Before you can use this function, you will need to add a reference to the System.Management.Automation.dll. This is not so straight forwarded because this .dll is a little hard to find. Take a look at this article that will help you add this reference.

https://www.codeplex.com/wikipage?ProjectName=ShinyPower&title=Frequently%20Asked%20Questions

Here is link for reference on the System.Management.Automation namespace:

https://msdn.microsoft.com/en-us/library/system.management.automation(VS.85).aspx

OK.. let's take a look at the function:

private Collection<PSObject> InvokePowerShell(string psscript)

{

RunspaceConfiguration rconfig = RunspaceConfiguration.Create();PSSnapInException Pwarn = new PSSnapInException();

       rconfig.AddPSSnapIn("Microsoft.SystemCenter.VirtualMachineManager", out Pwarn);

       Runspace runspace = RunspaceFactory.CreateRunspace(rconfig);

       runspace.Open();

       Pipeline pipeline = runspace.CreatePipeline();

       pipeline.Commands.AddScript(psscript);

       Collection<PSObject> results = new Collection<PSObject>();

       try

       {

            results = pipeline.Invoke();

       }

       catch(Exception ex)

       {
       }
      
       finally
       {
             runspace.Close();
             runspace.Dispose();
       }
}

 In this post I'll describe how to request and work with SCVMM template items via powershell and C#.

This post is going to assume you know how to setup C# and work with the System.Management.Automation namespace. If you don't I'll be posting a blog on how I do this next.

In this blog I'm going to show you how to request a specific template and parse the PSObject returned so we can populate a form with the template values. 

The first thing I'm going to do is create the powershell script to get a given template. I've created a function that takes in two parameters, the name of the template that I want, and the VMM Server where that template lives:

public void GetTemplate(string TemplateName, string Server)

{

string psscript = "Get-Template -VMMServer " + Server + " | where {$_.Name -eq '" + TemplateName + "'}";

      
      Collection<PSObject> results = InvokePowerShell(psscript);
}

The InvokePowerShell function will be detailed in the next blog post.

OK this is great, now we have a PSObject that contains all the members of the Template class. But how do we go and get these members so we can use them in C#? With the exception of the custom properties, you just have to know the Member name. Let's take a look:

if (results[0].Members["FullName"].Value != null)

{

tempTemplate.FullName = results[0].Members["FullName"].Value.ToString();

}

So here I am assuming that there is only one template that matches the name that we passed into our function. If there are multiple template objects, they will be returned via an array of PSObjects. This is why I just check the first item in the results array.

From there I just need to know the name of the member of the template class that I'm looking for. There are many ways to do this. You can open the SCVMM powershell window and type Get-Help Get-Template. Or easier, just open the SCVMM powershell window and type get-template. This will return all templates and all members of those templates of the localhost.

Now I did mention that the custom properties are a bit different to work with. This is because the custom properties are canacatenated together split by a space. We can create an array of custom properties to access our custom properties very easily like this:

string[] customproperties = (string[])results[0].Members["CustomProperties"].Value;

tempTemplate.CustomProperty1 = customproperties[0];

string CustomProperty2 = customproperties[1];

The last topic to cover would be setting template settings. I do this by getting the template from SCVMM and then passing the values of the members to be set via the  Set-Template cmdlet. This function will set custom property 1 of the template to the value passed into the function.

public void UpdateTemplate(string TemplateName, string myValue)

{
     string psscript = "";

     psscript += "\r\n$Template = Get-Template -VMMServer localhost | where {$_.Name -like '" + TemplateName + "'}\r\n";

     psscript += "\r\nSet-Template -Template $Template -Custom1 '" + myValue + "'";

     Collection<PSObject> results = InvokePowerShell(psscript);
}

If the function succeeds it will return the name of the template in the PSObject results.

Thats it... pretty straight forward.

Enjoy!

Adam