Monthly Archive

Categories

Windows Server 2012

New PowerShell console on Server Core

Server Core is great for reducing the footprint of your VMs – Nano server is smaller but it can’t be a domain controller

 

One draw back to server core is that you only get a single console. If you hang that for any reason you have to either try and open another one (Hyper-V console greys out CTRL-DEL-ALT) or open a few when you logon to the machine.

 

You still get a cmd.exe console instead of PowerShell – that should be changed. Its 10 years since PowerShell came along! So run Powershell to open  Powershell in the default console.

 

"Start-Process -FilePath powershell.exe -Verb RunAS" > new-powershell.ps1

Will create a simple script to open a new elevated Powershell console .

 

Run it as many times as you want. Perform your work in the new Powershell console and if it hangs – just shut it down. Keep the default console for just opening new PowerShell consoles and then you’ll always be able to keep working.

Don’t reinvent the wheel

Way back when I used to take Microsoft certification exams there were often questions of the form “Perform task X with the minimum of administrative effort” Most, if nor all, of the possible answers would be correct but the correct answer was the one that achieved the goal with the minimum amount of work.

 

Many, if not most, administrators don’t seem to follow that model.

 

This was brought home to me when I saw a forum discussion about collecting event log information from a bunch of remote servers on a regular basis.

 

You could set up a scheduled task/job that runs a script against the remote servers – collects the  log information and populates an Excel spreadsheet

OR

You could enable event log forwarding and just interrogate the combined logs as needed.

 

The second option is the easier to MAINTAIN and will cost you less effort in the long run.

 

When you start to solve a problem – stop and search for a bit to see if there is a solution already available in Windows server. Bet you’ll be surprised by what you find

DHCP scope lease time

I wanted to reduce the lease time on a DHCP scope

 

$lt = New-TimeSpan -Hours 12
Set-DhcpServerv4Scope -ScopeId 10.10.54.0 -LeaseDuration $lt

 

You could even make it a one liner if you wished

 

Set-DhcpServerv4Scope -ScopeId 10.10.54.0 –LeaseDuration (New-TimeSpan -Hours 12)

Formatting disks – – the new way

Last time I showed how to format disks using the Win32_Volume CIM class. If you need to perform this activity on a Windows Server 2012/Windows 8 or later system you can use a couple of cmdlets from the Storage module

Get-Volume | where DriveLetter -ne 'C' | Format-Volume -FileSystem NTFS -Confirm:$false –WhatIf

 

If you’ve not looked at the Storage module before there is a lot of useful cmdlets.

Working with Server Core–renaming the machine

When you create a new Windows server it usually assigns a name of its own. I always prefer using my own names for servers – I usually create the name so it gives some indication of the machine’s purpose.

Renaming a computer is simple

On the local computer run

Rename-Computer -NewName test01 -Force –Restart

This will rename the computer and force the restart.

If you want to do this remotely the cmdlet has a –ComputerName parameter plus the ability to define local or domain credentials.

Working with Server Core–setting IP addresses

When you create a new Windows machine it defaults to using DHCP to get an IP address. While that is fine for client machines most people apply a static address to their servers. Up until Windows 2012 you had 2 choices – use the GUI or use PowerShell and WMI.

Server 2012 introduced a host of cmdlets for administering your network settings.  Setting an IP address is simple as this:

New-NetIPAddress -InterfaceIndex 12 -IPAddress '10.10.55.101' -AddressFamily IPv4 -PrefixLength 24

 

I haven’t used it but you can also set the default gateway which would make the command

New-NetIPAddress -InterfaceIndex 12 -IPAddress '10.10.55.101' -AddressFamily IPv4 -PrefixLength 24 -DefaultGateway '10.10.55.01'

 

The DNS server addresses can be set like this

Set-DnsClientServerAddress -InterfaceIndex 12 -ServerAddresses '10.10.55.100'

 

The cmdlets are from the NetTCPIP and DnsClient modules respectively.

THESE MODULES ARE ONLY AVAILABLE ON WINDOWS 8/2012 AND LATER.

Working with Server Core–Domain join

Windows Server can be installed in two ways – full fat GUI or Server Core.  The latter is Windows without Windows.  The GUI components are stripped off and you’re just left with the core components.  This results in a smaller server – I’ve got 2 machines running in my Hyper-V test environment. Both are set to use dynamic memory with 512MB startup RAM. The full GUI machine needs 1126MB while the Server Core version needs 566MB.

With Server Core all you get is a prompt for administration – unfortunately its cmd.exe but typing powershell starts PowerShell – including running a profile.

Without a GUI you need to use the command line to do everything – I’ll be doing my demos at the PowerShell Summit from Server Core machines so some things will have to wait until after then – and I’m going to do a series on posts on administering Server Core machines.

I’ve already shown you  how to test if your machine is activated.  This how you join it to the domain.

Make sure that the IP and DNS server addresses have been set so the machine can find a domain controller.

$cred = Get-Credential
Add-Computer -Credential $cred -DomainName sphinx -Restart

Create a credential for the account that can join the machine to the domain.  use Add-Computer and supply the credential and domain name. The –ReStart parameter forces a restart post domain join.

if you want to see the results of calling Add-Computer then drop the restart switch and use Restart-Computer whenever you want the restart to happen

Checking license activation

I’m building some virtual machines for my demo’s at the upcoming PowerShell summit.  To make the demo’s, and setup, more interesting(?) I decided to use some Server Core instances.

The usual setup activities become a bit more interesting with Server Core – particular Windows activation. 

Windows 2012 R2 will activate itself if the new machine has an Internet connection when it is created. With the GUI version of Windows you can check that Windows is activated using the System applet in Control Panel.

If you’re using Server Core you can use WMI to test activation:

Get-CimInstance -ClassName SoftwareLicensingProduct |
where PartialProductKey |
select Name, ApplicationId, LicenseStatus |
Format-List *

Use the SoftwareLicensingProduct WMI class and filter for PartialProductkey  - that means a product key has been entered. You can then select the name of the product the ApplicationId and the LicenseStatus:

Name          : Windows(R), ServerStandard edition
ApplicationId : 55c92734-d682-4d71-983e-d6ec3f16059f
LicenseStatus : 1

A License status of 1 indicates that its licensed – i.e. activated

More on using WMI to test and set activation in chapter 13 of PowerShell and WMI – www.manning.com/siddaway2

VM disk info

A question came into the forum about getting information on the virtual disks associated with particular Hyper-V virtual machines. Is a bit of a digging exercise but this gets the results:

Get-VM |
foreach {
$VMname = $psitem.Name
Get-VMHardDiskDrive -VMName $VMname  |
foreach {
   Get-VHD -Path $_.Path |
   select @{N='VMname'; e={$VMName}}, Path, Size, FileSize
}
}

Get the set of VMs and for each of them get the VMHardDisks associated with the machine. For each VMHardDisk get the VHD – this is where you find the data you need.

 

The size and file size are in bytes – its a simple matter to create calculated fields that divide by 1GB or percentages if you require

AD Month of Lunches–Chapt 18 & 19 in MEAP

An updated MEAP has been released for Active Directory Management in a Month of Lunches.  This one adds chapters 18 & 19

  • Chapter 18, "Managing AD trusts"
  • Chapter 19, "Troubleshooting your AD"

The MEAP is available from www.manning.com/siddaway3

Enjoy