msgoodies

Information Overload and Social Impact of UC

2010-02-09T10:26:00.001+01:00

Michael J. Killian has created a 3-series post on his view on Unified Communications, Information Overload, the Social Behavior Impacts and Social Networking in The Enterprise.

Interesting and well written perspectives for anyone interested in the non-technical side of working with Unified Communications!

NET acquires SmartSIP from Evangelyze

2010-02-04T17:02:00.001+01:00

Good news for users of NET Gateways for their OCS implementations (and future customers ;-) NET has acquired the smartSIP product from Evangelyze -

The SmartSIP product line includes both the SmartSIP and SmartVoIP products. SmartSIP enables UC presence and interoperability for SIP phones while providing integration for Microsoft's Office Communications Server with IP-PBXs and integration for ITSP voice providers. SmartVoIP, a 2008 Internet Telephony Product of the Year, is an extension of SmartSIP that provides branch office integration with Microsoft Office Communications Server.

Read the full press release here.

Invoking the PowerShell Debugger from Script

2010-02-04T13:40:00.001+01:00

With PowerShell v2, a new and much improved command line debugger was introduced. The old one is still around though. Anyway, more information can be found in the help subject about_Debuggers.

The strange part is that Set-PsDebug –Step invokes the old debugger and there does not seem to be a way of invoking the new one. You can only invoke the new one by setting a breakpoint. Even though, breakpoints are a very useful feature which I use a lot, I would also like to do it from inside a script.

I have played around with some ways of doing this.

First, a self-contained function

Function Invoke-Debugger{
    function debug{}
    $bp=Set-PSBreakPoint -Command debug
    debug
    $bp | Remove-PSBreakpoint
}

function test{
    write-host 1
    Invoke-Debugger
    write-host 2
}

test

It works great, but has the downside, that the current execution pointer is inside the Invoke-Debugger function -

1
Entering debug mode. Use h or ? for help.

Hit Command breakpoint on 'debug'

x.ps1:4       debug
7 $docs>>> l

    1:  Function Invoke-Debugger{
    2:      function debug{}
    3:      $bp=Set-PSBreakPoint -Command debug
    4:*     debug
    5:      $bp | Remove-PSBreakpoint
    6:  }
    7:
    8:  function test{
    9:      write-host 1
   10:      Invoke-Debugger
   11:      write-host 2
   12:  }
   13:
   14:  test

No matter how I try to tweak it, I end up in the same way.

Next, lets try using a two part approach (setting the breakpoint and doing some action to invoke it) -

$null=Set-PSBreakpoint -Variable InvokeDebugger

function test{
    write-host 1
    $InvokeDebugger=1
    write-host 2
}

test

This is much better, now the execution pointer is right in the code -

1
Hit Variable breakpoint on '$InvokeDebugger' (Write access)

x.ps1:5       $InvokeDebugger=1
9 $docs>>> l

    1:  $null=Set-PSBreakpoint -Variable InvokeDebugger
    2:
    3:  function test{
    4:      write-host 1
    5:*     $InvokeDebugger=1
    6:      write-host 2
    7:  }
    8:
    9:  test
   10:

Eventually, this led me to this piece of code. It is easier to write than the variable assignment and you can also define an easy-writeable alias for it -

function Invoke-Debugger{}
New-Alias id Invoke-Debugger
$null=Set-PSBreakPoint –Command Invoke-Debugger

function test{
    write-host 1
    id
    write-host 2
}

test

The execution pointer is right at the call. If you include any statements in Invoke-Debugger, this will not work as well while ‘step’ will take execution into the function -

1
Hit Command breakpoint on 'Invoke-Debugger'

x.ps1:9       id
13 $docs>>> l

    4:  New-Alias id Invoke-Debugger
    5:  $null=Set-PSBreakPoint –Command Invoke-Debugger
    6:
    7:  function test{
    8:      write-host 1
    9:*     id
   10:      write-host 2
   11:  }
   12:
   13:  test
   14:

This method also enables you to make conditional break using straight, normal code (compared to making the logic in the –action argument of Set-PSBreakPoint) -

filter test{
    write-host "got $_"
    if ($_ -eq 3) {id}
}

1..5 | test

and the output -

got 1
got 2
got 3
Hit Command breakpoint on 'Invoke-Debugger'

x.ps1:9       if ($_ -eq 3) {id}
14 $docs>>> l

    4:  New-Alias id Invoke-Debugger
    5:  $null=Set-PSBreakPoint –Command Invoke-Debugger
    6:
    7:  filter test{
    8:      write-host "got $_"
    9:*     if ($_ -eq 3) {id}
   10:  }
   11:
   12:  1..5 | test
   13:

You can include the Invoke-Debugger function and the Set-PSBreakPoint in your profile, so they are available in all our scripts.

Happy debugging..

One stop shop for all your OCS patches

2010-01-29T09:18:00.001+01:00

Microsoft has created a really useful Updates Resource Center with all the latest and greatest patches listed for both Office Communications Server and the clients.

You can also find the updates at this RSS feed.

Note this is meant as a One Stop Shop, it is NOT the place where you’ll be notified first on a new OCS/OC patch.

Good work Microsoft ;-)

Interesting webcast from AudioCodes on OCS integration futures

2010-01-25T11:32:00.001+01:00

Besides not being totally honest on what “other vendors” can do (E.g. NET and others provides most of these features and more on top), there were a lot of different interesting news -

They spoke about Direct Connect (Formerly known as “Advanced Gateway” functionality) that apparently is supported in all their gateways from MP11x series to the Mediant (Se their Application Note that doesn’t detail QoE / ICE support however). The good thing about direct connect is that you get rid of the need to buy and deploy Mediation Server’s (Bad news are that it’s officially unsupported – so AudioCodes / NET has to support these configurations directly)

Other more interesting functionality was support for third party PBX handsets (Again like NET VX1200 series) on remote locations and survivability for these (If they support it). Also their own AudioCodes endpoints support registration to OCS, by proxying through their gateways to OCS (Which then handles call control). This includes support for Wideband and future support for presence.

Also Still time to see the #AudioCodes US prez running today at 12:00 noon (EST) / 09:00am (PST) - register here http://bit.ly/8LrYlL. And a recorded webcast should also become available soon (I will update link to it).

Comments on the OCS 2007 R2 Workload Architecture Poster

2010-01-24T01:36:00.002+01:00

The recently RTW’ed Architecture Poster provides a very good overview of port and certificate requirements in the different OCS workloads.

This poster of Office Communications Server 2007 R2 describes the traffic flow of protocols and ports used in each workload. Communications Server 2007 R2 supports the following workloads: IM and Presence, Conferencing, Application Sharing, and Enterprise Voice. These filtered views can assist you in architecting your deployment of Communications Server 2007 R2. The different server roles are described along with server certificate requirements. Firewall and DNS configuration requirements are also described.

I like this Poster and the idea/work put in to it and will certainly print one out for the walls in my home office. It provides a visually good overview of the Port usage and signaling/media flows used in OCS.

I have a few comments to the drawing though -

Application Sharing Workload

Red arrow depicting RDP/SRTP shows inbound traffic to 50,000-59,999. This is not correct – only outbound is required to endpoint. The only place this would be required is for traffic to an OCS 2007 “R1” Edge Server.
“A/V Edge must have publicly routable IP addresses” – true if implemented in loadbalanced config as shown (But not required for standalone Edge).

Enterprise Voice Workload

Red arrow depicting RDP/SRTP shows inbound traffic to 50,000-59,999. This is not correct – only outbound is required to endpoint. The only place this would be required is for traffic to an OCS 2007 “R1” Edge Server
I’m sure G.711 is not used through the A/V Edge as any packet loss would kill it ;-) Siren maybe used for conferencing scenarios.

A/V and Web Conferencing Workload

Arrows for HTTPS traffic are not correct – they should point towards the LM endpoints, as they are used for downloads of content e.g. slides.

Firewall configuration and ports on the Edge Server. Even though not OCS specific I would personally add port 53 for DNS (To internal or external depending on config) and port 80 to both external and internal (As this port is used for CRL checks). If not in the drawing then in the “Firewall Configuration” text box.

I generally like the idea about the DNS and Certificate portion in this poster, but IMHO it is to simplified. If a future update is planned the I think it should have its own page/poster to really handle the different scenarios and namespace requirements in OCS. So for certificates I would still point to the Whitepaper on Deploying Certificates in OCS 2007 and OCS 2007 R2, which does a better job of explaining the complexity of certificates usage/naming in OCS.

BETA Exam 71-404 OCS 2007 R2 – UC Voice Specialization, Will Be Available January 25, 2010

2010-01-22T08:54:00.001+01:00

The Voice Specialization exam for OCS 2007 R2 is finally getting closer to release with the launch of the Beta Exam.

Exam 74-404 (Beta Exam 71-404) is designed to validate the skills and experience of technical professional to design, deploy, and administer Microsoft Voice solutions in production based on Microsoft® Office Communications Server 2007 Release 2 (OCS 2007 R2) and including interoperability with OCS 2007, Microsoft® Exchange Server 2007 Unified Messaging, Microsoft® Windows Server Active Directory, gateways and Private Branch Exchanges (PBX’s).

It is highly recommended that candidates for Exam 74-404 have experience with OCS 2007 R2 and have completed the Unified Communications Voice Ignite v 2.0 (R2) Workshop. A preparation guide for the exam will be available soon. Click here for more details.

Do remember that Inceptio Learning Solutions provide fully qualified UC Voice Ignite Workshops on-site at your company/training provider (almost) anywhere in the world (Contact my.initials@inceptio.dk for further info).

Happy testing ;-)

Maybe I should redefine “Office 2010 x64 fully working with OC” ?

2010-01-20T13:01:00.001+01:00

I have now been working with Office 2010 x64 a week or so as per my post Office 2010 x64 now fully working with OC and LiveMeeting Add-in. I have some intermittent problems though – when I log-in everything works fine, then on occassion I get an -

The problem appears after a while and then disappears again later. Integration seems to work fine except that this problem appears from time to time – did anyone get the same error or no error at all? (Perhaps with only one Exchange mailbox – I have two Exchange mailboxes which has other issues – so it may be part of this problem).

Reference Variables

2010-01-18T16:20:00.001+01:00

If you want to change a value inside a function, you have to use reference variables. They do not work the same way as they do in C, C# or VbScript.

This is a simple example -

function a([ref]$v) {
    "a1 v=$($v.value)"
    $v.value++
    "a2 v=$($v.value)"
}

$x=0
"begin x=$x"
a ([ref]$x)
"end x=$x"

Note that must use [ref] when declaring the argument and also when calling. Note also that you must use ([ref]$x) in the call. Finally, you can see that [ref] actually converts the variable into a PSReference object and that you must use .Value to set/get the value.

But what if you want to transfer the reference variable across multiple function calls? Well, the logical approach is this -

function b([ref]$v) {
    "b1 v=$($v.value)"
    $v.value++
    "b2 v=$($v.value)"
}

function a([ref]$v) {
    "a1 v=$($v.value)"
    $v.value+=10
    b ([ref]$v)
    "a2 v=$($v.value)"
}

$x=0
"begin x=$x"
a ([ref]$x)
"end x=$x"

The pattern is simply repeated. Just like you would in other languages. But THIS WILL NOT WORK. As the first call creates a PSReference object, using ([ref]$v) creates another level of redirection and things will fail. The correct solution is -

function b([ref]$v) {
    "b1 v=$($v.value)"
    $v.value++
    "b2 v=$($v.value)"
}

function a([ref]$v) {
    "a1 v=$($v.value)"
    $v.value+=10
    b $v
    "a2 v=$($v.value)"
}

$x=0
"begin x=$x"
a ([ref]$x)
"end x=$x"

Note that in the second call, $v – a PSReference object – is simply transferred.

Bottom line: Avoid [ref] when you can. Return values or use scoped variables when possible. Remember that it is very easy to return multiple values from a function and assign them to different variables -

function c{
    "Per"
    "Denmark"
    44
}
$Name,$Country,$ShoeSize=c

Simple script for extracting CDR records from OCS Monitoring Server

2010-01-16T19:25:00.001+01:00

Andre Blumberg created instructions and a small script for extracting CDR records for outbound calls. Find it here and do remember to provide your feedback for him (So we all can benefit from your/Andre’s enhancements to the script ;-)

Office 2010 x64 now fully working with OC and LiveMeeting Add-in

2010-01-15T13:14:00.001+01:00

First part of the solution was the January Office Communicator Update to 6907.83 that solved the Outlook integration issues/error (Which forced me to “downgrade” to 32 bit Office).

The last part of this solution was the release of the 64-bit LiveMeeting add-in which can be found here and while you’re at it the update to the LiveMeeting console can be found here.

Before upgrading do remember that there is a problem with file uploading in LiveMeeting – only supported file format is native PowerPoint files. Also remember the workaround to make updating OC running 64 bit OS work correctly.

Happy installing/upgrading to 64 bit ;-)

Workaround for Client Version Filtering problem using 32 bit OC on X64 OS

2010-01-13T21:30:00.001+01:00

My colleague Claus-Ole also found and fixed the problem, but Michael Sneeringer (A OCS MCM) was kind enough (hint, hint) to describe the problem and solution in Client Version Filtering on Windows x64.

OCS Wave 14 features

2010-01-07T08:10:00.001+01:00

FYI - Curtis Johnstone has revealed some of the essential OCS Wave 14 features – find them in his post OCS in 2010 - The UC 14 Wave.

As an MVP I’m under NDA so I can’t add to the above, but what I can say is that the future for OCS looks promising ;-)

MS PM comments on CUCiMOC(kup) and the joint Cisco / Microsoft support statement

2009-12-08T14:44:00.001+01:00

Take a look at the blog post Cisco: Just Like Any Other Office Communications Server ISV? for some official words (finally) from Microsoft on Cisco’s CUCiMOC and Microsoft Office Communications Server integration.

He hits the nail on the many questions I have raised at customers, including the supportability problems in terms of integration at the desktop level with program version dependencies and especially the (lost) user experience when using CUCiMOC. Furthermore Currently Cisco is setting a Version 8.0 / Q3 CY10 date for supporting Windows 7 (And also with limited support for 64 bit).

Furthermore (as he also notes) it is important to note the dependencies and lost features (RDP Sharing, Audio/Video/Webconferencing etc – that you would need to buy from Cisco).

Read the joint support statement here and Haberkorns blog post for more info.

Messing with output from Format-Table

2009-11-19T21:26:00.001+01:00

Found at question at psobject.codeplex.com:

I was wondering if I could write the outout without any blank spaces between 2 fields. For example, I am using hash tables to display 2 columns from dir output. But it always comes up with a blank space between those 2 data elements. I need this to generate a fixed format output with data elements only and no spaces in between. Any help is greatly appreciated.

$column1 = @{expression="mode";width=5;label="mode";alignment="left"}
$column2 = @{expression="name";width=10;label="name";alignment="left"}

$dir |format-table $column1,$column2

$mode name
---- ----
d---- download
d---- extract
-a--- alias.txt
-a--- Compute...
-a--- execute...
-a--- get_dn.ps1
-a--- hh

Well, it can be done. I looked into the objects the Format-Table spit out and after some poking around, I came up with this -

$column1 = @{expression="mode";width=5;label="mode";alignment="left"}
$column2 = @{expression="name";width=20;label="name";alignment="left"}

# Save widths, all non-fixed length value should specify a width
$widths=@{}
dir | format-table $column1,@{l="|";e={"|"}},$column2 | foreach {
    if ($_.pstypenames[0] -eq "Microsoft.PowerShell.Commands.Internal.Format.FormatEntryData") {
        # Capture the values and convert them to one value
        $value=""
        $count=$_.formatentryinfo.formatPropertyFieldList.count
        foreach($i in 0..($count-1)) {
            $value+=$_.formatentryinfo.formatPropertyFieldList.item($i).propertyvalue.ToString().padright($widths.$i)
        }
        # Delete all but one field
        $_.formatentryinfo.formatPropertyFieldList.removerange(1,$count-1)
        # and update its value
        $_.formatentryinfo.formatPropertyFieldList.item(0).propertyValue=$value
        $_
     }
     elseif ($_.pstypenames[0] -eq "Microsoft.PowerShell.Commands.Internal.Format.FormatStartData") {
        # Capture the headers and convert them to one header
        $value=""
        $width=0
        
        $count=$_.shapeinfo.tablecolumninfolist.count
        foreach($i in 0..($count-1)) {
            $w=$_.shapeinfo.tablecolumninfolist.item($i).width
            $width+=$w
            $widths.$i=$w
            $value+=$_.shapeinfo.tablecolumninfolist.item($i).propertyname.ToString().padright($w)
        }
        # Delete all but one field
        $_.shapeinfo.tablecolumninfolist.removerange(1,$count-1)
        # and update its value
        $_.shapeinfo.tablecolumninfolist.item(0).propertyName=$value
        $_.shapeinfo.tablecolumninfolist.item(0).width=$width
        $_
     }
     else
     {
        $_
     }
}

If you like it, convert it to a function as en exercise ;)

Happy formatting!

PS Remoting to Home Server

2009-11-19T16:25:00.001+01:00

I wanted to PowerShell remote to my home server, but as it is – for the good reason, that it is impossible – not in the domain of my PC, I have to add it to TrustedHosts.

This is my story.

First, I enabled PS remoting on the home server with a simple

Enable-PSRemoting

Next, I attempted to access the server

enter-pssession server -cre server\administrator

By doing so, I received a very long error message.

I tried with –authentication negotiate. This reduced the error message to 10 lines ;) It told me to configure TrustedHosts with Winrm.cmd.

I looked at Winrm.cmd, but that looked very complicated and this is at the end of the day. Luckily, the WSMAN drive popped up in my mind. I switched to my administrative account and did this

PS C:\Users\user> cd wsman:
PS WSMan:\> dir

   WSManConfig:

ComputerName                                  Type
------------                                  ----
localhost                                     Container

PS WSMan:\> cd .\localhost
PS WSMan:\localhost> dir


   WSManConfig: Microsoft.WSMan.Management\WSMan::localhost

Name                      Value                                                             Type
----                      -----                                                             ----
MaxEnvelopeSizekb         150                                                               System.String
MaxTimeoutms              60000                                                             System.String
MaxBatchItems             32000                                                             System.String
MaxProviderRequests       4294967295                                                        System.String
Client                                                                                      Container
Service                                                                                     Container
Shell                                                                                       Container
Listener                                                                                    Container
Plugin                                                                                      Container
ClientCertificate                                                                           Container


PS WSMan:\localhost> cd .\Client
PS WSMan:\localhost\Client> dir


   WSManConfig: Microsoft.WSMan.Management\WSMan::localhost\Client

Name                      Value                                                             Type
----                      -----                                                             ----
NetworkDelayms            5000                                                              System.String
URLPrefix                 wsman                                                             System.String
AllowUnencrypted          false                                                             System.String
Auth                                                                                        Container
DefaultPorts                                                                                Container
TrustedHosts                                                                                System.String


PS WSMan:\localhost\Client> Set-Item .\TrustedHosts server

WinRM Security Configuration.
This command modifies the TrustedHosts list for the WinRM client. The computers in the TrustedHosts list might not be authenticated. The
client might send credential information to these computers. Are you sure that you want to modify this list?
[Y] Yes  [N] No  [S] Suspend  [?] Help (default is "Y"): y
PS WSMan:\localhost\Client>

Back to my standard user

194 WSMan:\localhost> enter-pssession server -cre server\administrator
[server]: PS C:\Documents and Settings\Administrator\My Documents>

and it worked.

Happy remoting

Laissez-Faire Access Control

2009-11-17T21:44:00.001+01:00

Bruce Schneier have an abstract of a paper which claims that enabling users to get the access they need while auditing the access is better than a centrally controlled setup.

Read it for yourself – at least read the abstract.

PowerShell 2.0 is supported for Exchange 2007 SP2

2009-11-16T14:00:00.001+01:00

In case you wondered, the greatly improved PowerShell 2.0 can be installed on servers running Exchange 2007. But you must be running Service Pack 2.

BTW: You cannot find PowerShell v2 on www.microsoft.com/downloads. You have to go to http://support.microsoft.com/kb/968929.

Implementing OCS presence in Outlook Live Server (a.k.a. Exchange 2010 OWA)

2009-11-12T09:32:00.001+01:00

The tools required to implement OCS presence in Exchange 2010 OWA has been released

Microsoft Office Communications Server 2007 R2 Web Trust Tool

Microsoft Office Communications Server 2007 R2 Web Service Provider

And instead of me showing how to do it, I would recommend you checkout the blog post Implementing integrated OCS in Exchange 2010 from Chris and Robin’s Technology Blog.

One addition I have to their post is that the reference material for the OCS Web Service Provider can be found on the OCS TechNet site here.

Picking files with the mouse using Get-DroppedFile

2009-11-11T10:22:00.001+01:00

Sometimes (often?) it is just easier picking your files with the mouse. As long as the files are in one folder that is not that hard, but if you have files scattered all around that is tougher. Also right-clicking and copy-as-path is annoying.

To make this easier, I have created a small drop box function. A small transparent window will be shows and when you drop files on it, those files will be send to the output pipeline where you can do the rest of your processing.

I made the forms part using Visual C#. Relativily trivial. And Add-Type enabled me to embed it into my script. The hard part was making it async so that files would appear ín the output pipeline as soon as they were dropped. I had to resort to good old VB5-style DoEvents (just revealed my age, I guess). If you can come up with a non-polling solution, please let me know.

What it can be used for -

Testing scripts with different files
Move photos to a folder, converting them as they are moved
Renaming files
Compressing files
continue the list yourself

All tasks where you – the human – can make the decision about what to do with a file are relevant.

With PowerShell v2 being available on all platform, do I have to say, that this is a V2-only script?

Please, read the comments in the script for further information.

Get-DroppedFile.ps1


<#
.Synopsis
    Create a drop box window and output the files dropped to the pipeline
.Description
    Create a drop box window. When files are dropped, they are send to the output pipeline right away.
    Stop Get-DroppedFile by closing the window.
.Inputs
    None
.Outputs
    File names (-asText), IO.DirectoryInfo or IO.FileInfo objects
.Example
    Get-DroppedFile | Copy -destination e:\ -passthru | foreach { $x.attributes=$x.Attributes.ToString()+",readonly" } 
    Copy dropped files to e:\ and set the readonly bit
#>
param(
   [string]
   # The caption of the drop box
   $Caption,
   [switch]
   # Return file names (full path) as text
   $AsText,
   [switch]
   # Recurse directories, I.E. the folder itself is not returned, only its children
   $Recurse,
   [switch]
   # (Internal switch used to detect -sta invocation)
   $_InternalReinvoked)

# Create the script code - direct execution or SingleThreadedApartment is determined later
$script={

$loaded=$false
try {
    $null=[system.type] "Get_DroppedFile.Form1"
    $loaded=$true
}
catch {}

# The form code. Created in Visual C# 2008 Express and slightly adopted
if (!$loaded) {
    add-type -TypeDefinition @'

using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Drawing;
using System.Windows.Forms;

namespace Get_DroppedFile
{
    public partial class Form1 : Form
    {
        Color defaultColor;
        public Form1()
        {
            InitializeComponent();
            defaultColor = this.BackColor;
            
        }
        
        public void avoidWarning()
	{
	}
      
        private void Form1_DragDrop(object sender, DragEventArgs e)
        {
            // Back to default color
            this.BackColor = defaultColor;
        }

        private void Form1_DragOver(object sender, DragEventArgs e)
        {
            // Start dragdrop
            e.Effect = DragDropEffects.Copy;
        }

        private void Form1_DragEnter(object sender, DragEventArgs e)
        {
            // visual feedback in drag
            this.BackColor = Color.FromArgb(defaultColor.ToArgb() - 0x101010);
        }

        private void Form1_DragLeave(object sender, EventArgs e)
        {
            // restore color
            this.BackColor = defaultColor;
        }
    }
}

namespace Get_DroppedFile
{
    partial class Form1
    {
        /// <summary>
        /// Required designer variable.
        /// </summary>
        private System.ComponentModel.IContainer components = null;

        /// <summary>
        /// Clean up any resources being used.
        /// </summary>
        /// <param name="disposing">true if managed resources should be disposed; otherwise, false.</param>
        protected override void Dispose(bool disposing)
        {
            if (disposing && (components != null))
            {
                components.Dispose();
            }
            base.Dispose(disposing);
        }

        #region Windows Form Designer generated code

        /// <summary>
        /// Required method for Designer support - do not modify
        /// the contents of this method with the code editor.
        /// </summary>
        private void InitializeComponent()
        {
            this.SuspendLayout();
            // 
            // Form1
            // 
            this.AllowDrop = true;
            this.AutoScaleDimensions = new System.Drawing.SizeF(6F, 13F);
            this.AutoScaleMode = System.Windows.Forms.AutoScaleMode.Font;
            this.BackColor = System.Drawing.SystemColors.ActiveCaption;
            this.ClientSize = new System.Drawing.Size(116, 49);
            this.Cursor = System.Windows.Forms.Cursors.Default;
            this.MaximizeBox = false;
            this.MinimizeBox = false;
            this.Name = "Form1";
            this.Opacity = 0.75;
            this.ShowIcon = false;
            this.Text = "Drop Box";
            this.TopMost = true;
            //this.Load += new System.EventHandler(this.Form1_Load);
            this.DragLeave += new System.EventHandler(this.Form1_DragLeave);
            this.DragDrop += new System.Windows.Forms.DragEventHandler(this.Form1_DragDrop);
            this.DragEnter += new System.Windows.Forms.DragEventHandler(this.Form1_DragEnter);
            this.DragOver += new System.Windows.Forms.DragEventHandler(this.Form1_DragOver);
            this.ResumeLayout(false);

        }

        #endregion


    }
}

'@ -verbose -ReferencedAssemblies system.drawing,system.windows.forms
}

# Create our form object
$form=New-object get_droppedfile.form1

# Get rid of any events leftover
get-event get-droppedfile -erroraction silentlycontinue | remove-event

# Add handlers
# The handlers transfer the action/file to the main loop using event
$form.add_dragdrop( { $null=new-event -sourceidentifier get-droppedfile -messagedata $args[1].Data.GetData("FileDrop", $true)  })
$form.add_formclosed( { $null=new-event -sourceidentifier get-droppedfile -messagedata "[close]" })

#Register-ObjectEvent -InputObject $form -EventName dragdrop #-SourceIdentifier blah#
#Register-ObjectEvent -InputObject $form -EventName formclosed  #-SourceIdentifier blah

# Custom caption
if ($caption) {$form.text=$caption}

# This is the tricky part. Dropping is quite simple, but feeding the output pipeline with
# the dropped files (so you do not have to wait until the drop box is closed) is not simple.
# I came up with this solution:
# - Do not use ShowDialog as is modal and will suspend PowerShell processing
# - use Show and doevents in a loop
# - this method consumes some CPU, but the Start-Sleep keeps it to a few per cent
# - The Event actions generates events and they are read here in the main loop
#   and converted to file names which are sent to the pipeline

# Show the drop box
$form.show()

do {
  $e=get-event get-droppedfile -erroraction silentlycontinue # suppress no such events
	$exit=$e.messagedata -eq "[close]" # Test close message
	if ($e.messagedata -and !$exit) {$e.MessageData} # Send file to pipeline
	if ($e) {$e | remove-event} # Remove event from queue
	if ($exit) {break}
	start-sleep -m 100 # Wait a little
	[System.Windows.Forms.Application]::doevents() # React to form events so the windows can be moved etc.
} while($true)

# Shutdown
$form.close()
} # end of script assignment


# Generate command for -sta recursive call. Done here where $myinvocation has the right value
$command = "&'" + $myinvocation.mycommand.definition + "' -caption '$caption' -_InternalReinvoked"

# Execute the next in a scriptblock so output can be piped
&{

	# Forms must run in SingleThreadedApartment style
	# Re-invoke PowerShell if necessary
    if ($host.runspace.ApartmentState -ne "sta") {


     write-verbose "Invoking PowerShell with -sta"
        $bytes = [System.Text.Encoding]::Unicode.GetBytes($command)
        $encodedCommand = [Convert]::ToBase64String($bytes)
        powershell.exe -sta -noprofile -encodedCommand $encodedCommand
     #powershell -sta -noprofile $script
    }
    else {
     &$script
    }

} | where {$_} | foreach {
    # Handle the different return options
	if ($_InternalReinvoked.ispresent) {
        # -sta call, always return strings
		$_
	}
	elseif ($recurse.ispresent -and (test-path -pathtype container $_)) {
        # Recurse folder tree, return text or objects
		Get-ChildItem $_ -recurse -force | foreach {
			if ($astext.ispresent) {
				$_.fullname
			}
			else {
				$_
			}
		}
	}
	elseif ($astext.ispresent) {
		$_
	}
	else {
		Get-Item $_
	}
}

Have fun

ATE Schedule for TechEd Berlin

2009-11-08T20:12:00.001+01:00

I’m in Copenhagen waiting for my flight to Berlin. If you want to touch base then I have booth duty at the UC Ask-The-Experts (and Dennis ;-) area Tuesday from 15:15 – 18:15 and Friday 11:30 – 14:45.

CU there !

Code Contracts and Pex

2009-11-04T14:11:00.001+01:00

If your are coding, you should check out this Channel 9 video (11:30 minutes) where Manuel Fähndrich and Peli de Halleux show you how to specify code contracts and testing them using Pex directly in your coding environment.

Here’s a screen shot showing the Contract statements in the code and the Pex test runs below -

Happy coding

OCS Cumulative Server Update Installer

2009-11-03T23:28:00.001+01:00

One of the biggest serviceability issues with OCS 2oo7 R2 hotfixes, has been that you as an admin had a list of 17 (Seventeen – yes) different hotfixes that you had to apply to the correct server/role. This could be done by checking the 13 (Thirteen) different server roles and the list of updates that applied to them and then manually installing each and every of the necessary updates separately.

This “design issue” has now been mitigated with the “Cumulative Server Update Installer”, that basically checks each server, its roles and then suggest and applies the necessary hotfixes to your OCS Server (How hard can it be ;-).

So one installer (and one download, that contains all the hotfixes), that handles all servers/roles and it can be done either through GUI or scripted through Command Line.

You can download the installer from KB968802 that contains a list of all the updates (October Patches), a description of the process for updating and a download link.

At the download site you can download each update, but just scroll down to the download called “ServerUpdateInstaller.exe” and only download this. Do note that you should execute it from an empty folder as it extracts all the necessary updates for the relevant server role on execution (After installation has been completed it removes updates and only logs from each applied update is left).

Below is an example from ServerUpdateInstaller executed on a mediation server:

As you can see it both lists existing installed version, the new updated version and not least a link to the hotfix KB.

Do note that the July Database Update found in KB969834 isn’t installed automagically, this update has to be run manually!

Good work Microsoft (And happy updating to all of you ;-)

PowerShell V2 now Available on Older Operating Systems

2009-11-02T09:50:00.001+01:00

Now you can safely forget V1 and switch to the vastly superior V2. As of last week it is now available on -

Windows Server 2008 with Service Pack 2

Windows Server 2003 with Service Pack 2

Windows Vista with Service Pack 2

Windows Vista with Service Pack 1

Windows XP with Service Pack 3

In Windows 7 and WS08 R2 it is part of the package.

PS2 is part of Windows Management Framework which also includes WinRM and BITS4.

BTW: You need to install PowerShell 2 on server core. Read how in KB 976736.

Windows Server 2008 R2 Service and Virtual Accounts

2009-11-02T09:01:00.001+01:00

One of the best reasons for upgrading to R2, is the new account types for managing services. Changing user account passwords being used for running services, scheduled tasks and application pools are often a real pain and consequently, often being skipped. And wouldn’t it be nice if it was handled automatically like a computer account? Well, that is exactly what R2 offers.

Two new types of service accounts are available in Windows Server® 2008 R2 and Windows® 7—the managed service account and the virtual account. The managed service account is designed to provide crucial applications such as SQL Server and IIS with the isolation of their own domain accounts, while eliminating the need for an administrator to manually administer the service principal name (SPN) and credentials for these accounts. Virtual accounts in Windows Server 2008 R2 and Windows 7 are "managed local accounts" that can use a computer's credentials to access network resources.

Read the Service Accounts Step-by-Step Guide for more information.