I have just finalized a custom VNC MSI File and a ADM file.

Firstly the ADM is not fully managed – so beware if you add it then remove it.

I have not changed any Tight VNC source.  Just streamlined the installation.

Notes:

The typical installation – installs VNC server as a service.  No viewer is installed – this is perfect for workstations.

So a quiet install would achieve this.

Customization the installation allows you to add the viewer, as well as the ADM template. (%Windows%/Inf) directory. Plus some other ancillary vnc stuff.  Check the Adm for a description of the settings

FINALLY – THIS INSTALLER SETS THE DEFAULT PASSWORD TO BE BLANK (not empty) and sets the Autherization required to disabled…meaning no password is required to connect to the workstations it is installed on! The Authhosts settings restricts connection to the localsubnet only. As does the installed firewall entry for the server.  THESE SETTING MAY NOT SUITE YOUR ENVIRONMENT – USE ORCA TO TRANSFORM THE INSTALLATION IF NEEDED.

Check the Tight VNC site for an explanation of settings.

This file was created with WIX.

VNC Network Installation File (updated 18/4/09)

If you find this useful please let me know – I am debating whether to learn enough C to reverse engineer Tight VNC to allow it to be fully managed under Group Policy.

I had seen a number of scripts detailing how to send mail from Powershell v2.0, but the limitation that always seemed to appear was that they used SMTP to do so and didnt allow complex mail formation… So I investigated .Net to see if it was possible, low and behold I came accross the following Class….

System.net.mail Namespace

and then more specifically

System.net.mail.mailmessage Class

It was this mail message class I based the following script on

#Collect Events from That are errors and report
$A = Get-date -hour 0 -minute 0 -Second 0
$logfile = $env:Temp,"applog.html" -join ""
$logfile2 = $env:Temp,"syslog.html" -join ""
$SmtpClient = new-object system.net.mail.smtpClient
$SmtpServer = "smtp.domain.com.au"
$SmtpClient.host = $SmtpServer
$log2 = Get-EventLog -LogName System -After $a -EntryType Error
$log = Get-EventLog -LogName Application -After $a -EntryType Error
If (($log.count + $log2.count) -gt 0)
{
$MailMessage = new-object system.net.mail.Mailmessage -Args "system@domain.com.au","user@domain.com.au"
$MailMessage.From = "System Log <ystem@domain.com.au>"
$MailMessage.Subject = "Syslogs for $a"
If ($log.count -gt 0){
$log | convertto-html -Property index, TimeGenerated, InstanceID, Source, Message   -Title "Application Log for $a" > $logfile
$mailmessage.attachments.add($logfile)
}
if ($log2.count -gt 0){
$log2 | convertto-html -Property index, TimeGenerated, InstanceID, Source, Message   -Title "System Log for $a" > $logfile2
$mailmessage.attachments.add($logfile2)
}
$SmtpClient.send($mailmessage)
$mailmessage.finalize
$mailmessage.dispose()
$SmtpClient.finalize
If (Test-Path $logfile){Remove-Item -Force -Path $logfile}
If (Test-Path $logfile2){Remove-Item -Force -Path $logfile2}
}

So thats the entire script…now onto a breakdown

So we start with this

$logfile = $env:Temp,"applog.html" -join ""
$logfile2 = $env:Temp,"syslog.html" -join ""

Here I declare two objects containing two temporary files I intend to fill with information and then attach to an email. $env:temp simply takes the default temp directory path and appends it two the file name with a join. A new cmdlet is planned called new-path which I think will be a better alternative here.

$SmtpClient = new-object system.net.mail.smtpClient
$SmtpServer = "smtp.domain.com.au"
$SmtpClient.host = $SmtpServer

This section declares a new object as a net smtpclient – this object will send my mail a little later.

$log2 = Get-EventLog -LogName System -After $a -EntryType Error
$log = Get-EventLog -LogName Application -After $a -EntryType Error

Here I am getting some info to send my self  – in this case I am asking for any errors in the application and system logs that have occured today. I define a$ previously as

$A = Get-date -hour 0 -minute 0 -Second 0

Basically a time call.  Okay now I need to process my data a little do some checking to ensure I am not processing nothing and then send the mail.

If (($log.count + $log2.count) -gt 0)
{

This just checks that one of the logs actually returned some entries otherwise we are wasting our time and I dont need empty emails!

$MailMessage = new-object system.net.mail.Mailmessage -Args "system@domain.com.au","user@domain.com.au"
$MailMessage.From = "System Log <ystem@domain.com.au>"
$MailMessage.Subject = "Syslogs for $a"

Now this part is the message.   I created a new object based on the Mailmessage class.  You need to specify the from and to address’s as args as the class property of  “to” is readonly and if you tried to declare it like

$mailmessage.to ="user@domain.com"

You would get an error stating the property was readonly.

Now some more error checking to ensure  which log returned values.  Then only attach that result.

If ($log.count -gt 0){
$log | convertto-html -Property index, TimeGenerated, InstanceID, Source, Message   -Title "Application Log for $a" > $logfile
$mailmessage.attachments.add($logfile)
}

Here we see I parse the result to a html file.  It makes for a nice easy read.  Then I call the Add method of the mailmessage.attachment property and add the log file to the message.

The attachments property is actually a collection and you can use a single command to add more that 1 file as long as you delimit each file with a comma.

I repeated the process for log2

$SmtpClient.send($mailmessage)

And then I send the message

Now I need to clean up. This part is really important particularly if the script is executed in a continually running powershell session rather than a 1 off instance.  For two reasons.  The mail message class once you add a file locks it.  Even once the script ends the file will remain locked.  It will only unlock after the powershell instance ends.  Thats too late as you cant remove your temp files and clean up after yourself which is just rude!

So I need to call two methods, firstly I finalize mailmessage, I am not sure if this was actually neccessary but reading the technet docs it seemed the smart way to go.  Then I dispose of the mailmessage object.  This unhooks it and leaves the files that you added open to removal.  I think after going a bit of reading net objects have a habit of doing this and it is probably good policy to always dispose the objects once you are finished if its possble.

So her goes

$mailmessage.finalize
$mailmessage.dispose()
$SmtpClient.finalize
If (Test-Path $logfile){Remove-Item -Force -Path $logfile}
If (Test-Path $logfile2){Remove-Item -Force -Path $logfile2}
}

And we are done.  One last comment

I hate errors, so I test for the existence of each logfile before removing them …just to avoid errors in case I never created them.

Windows PowerShell 2.0 includes several significant features that extend its use, improve its usability, and allow you to control and manage Windows-based environments more easily and comprehensively. Windows PowerShell 2.0 is backward compatible. Cmdlets, providers, snap-ins, scripts, functions, and profiles that were designed for Windows PowerShell 1.0 work in Windows PowerShell 2.0 without changes.

At the outset I am going to state that using the standard windows (xp) removable drive drivers, you cannot partition a USB drive.  It just is not supported, you can reformat the drive to NTFS (for security) with a bit of tweaking, but you cant create multiple partitions.

So I found this simple trick to make windows see a USB as a fixed drive and allow multiple volumes and partition resizing.    You will need admin privileges for the system.

The first step in this process is to plug in the USB drive making sure it contains no critical files and just to be sure format it as Fat32 using the standard windows methodology.

Now determining the USB drives HID.  or Hardware ID, you are going to need this to tweak the install file for the new drivers.  To do this there are a number of methods. The easiest I have found is to open “Device Manager” .  Click the view menu and select “View by connection”. Now open the drop down lists below (the names may vary depending on youy system) until you find the PCI to USB Controllers (there may be more than one) open each in turn and until you see the name of the USB drive you have connected listed.  Below that should be “Generic Volume” Right click and select properties. Then select the details tab and in the “Property” drop down list select “Device Instance ID” – copy the displayed value to a notepad.

Alternatively if this does not work look in the registry under

HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Enum\USBSTOR

find your Drive and copy the key using everything after enum eg:

USBSTOR\DiskBUFFALO_USB_Flash_Disk__2.00

Cool you now have the device id.  Now you need to use it to edit the install (inf) file for the new driver.

I used the following driver “Hitachi Microdrive” Download it and save it to your desktop and extract the files to desktop.

One of the extracted files is called CFADISK.inf

Open it with your favorite text-editor I prefer Notepad++ found here.

Now find the section entitled [cfadisk_device]

Within that section you will find lines of text like this:

%Microdrive_devdesc% = cfadisk_install,IDE\DiskIBM-DSCM-11000__________________________SC2IC801

Basically what you need to do is comment out all (in this section) of them except one, the first one, place a “;” at the start of each line you wish to comment out.  In the first line replace anything after the “%Microdrive_devdesc% = cfadisk_install, “ with your Device Instance ID.  Now save the file.

Now you need to update the driver for your usb drive with your modified driver.  Go back to your Device manager and select the Generic Volume you found before (make sure its the right one) and chose “Update driver” from the driver tab.

On the first screen select “Not at this Time”

Second screen Select “Install from a Specific Location(advanced)

Third Select Dont search I will Choose the driver

Finally select “Have Disk”

Then Browse and chose the folder you saved the driver and install file you just updated in.

If you changed the ID correctly it should show you the a driver.

I have this working under XP but not under Windows 2003 I have yet to find a driver that 2003 will allow.

Keep in mind you need to install the driver on every system you wish to access anything above the first partition.  SO what I did is this

Created the first and largest partition as FAT32 I then created a Secure NTFS partition after this… This partition is not accessible unless I install the driver which I keep in the primary partition.  (copy it to desktop before you try and install it)

Good luck

Extra details here

http://www.pcreview.co.uk/forums/thread-2213125.php