In the third in this video series I wanted to show you NIC Teaming..
which is how you can provide a single network interface from multiple interface even if these are from different manufacturers. That's why in the video below I have used some Belkin USB Ethernet adapters combined with the on-board network card on my laptop.
If you want to try this yourselves you would install the hyper-V role on windows server 2012 possibly using the introduction to this series as a guide as well as the introduction to Hyper-V I published yesterday. You would then create several internal virtual switches in Hyper-V and then create a new virtual machine (VM) with several network cards in it bound to these virtual switches. There’s one property you’ll want to change in the settings for the VM so that NIC teaming works properly which is to set the network adapter in the VM to be used for NIC teaming:
The two key things to remember about NIC teaming from this are:
For further reading there’s a deployment guide you can download here
Finally if you haven’t got a TechNet subscription and want to try this yourselves you can get an Evaluation Edition of Windows Server 2012.
Having looked at how NIC teaming makes the best use of your network card in my last video I wanted to explore another networking feature in Windows Server 2012, DHCP Failover..
You might be wondering why this matters given that split scopes have been around for ages and you can also create a DHCP role in a windows cluster. The clustering option requires shared storage to work whereas in DHCP failover in Windows Server 2012 all you need is a shared secret (a password). Split scopes might mean you run out of IP address on one node and there is no high availability as such. Whereas in DHCP failover you can either split the IP address allocation, or use one of the server as a hot standby.
Given that you can only have setup two server in a DHCP Failover configuration like this I would see this as a great solution for smaller organisations who can’t justify running a cluster for this workload, but need some resilience for key services like DHCP.
If you want to try it do refer to parts 1 & 2 of this series to setup your demo rig. and for more details on DHCP failover you may also wish to check out TechNet here
The main reason my blog has been totally neglected for the last few months is because Simon and I have been on tour doing IT camps. Despite the number of these we have done there will be many of you of you who haven’t been able to attend these because of time constraints or because you simple didn’t realise that we were doing them. To correct this Simon and I are going to record some of the best demos we have done and publish these over the coming weeks. The series is called Evaluate This! For the simple reason that we want you to try some of this out yourselves just like we encourage you to do if you come along to camp.
Setting these up yourselves allows you to skill up, and work out how and if these new features will work in your organisation. Trying out these new features will also help you prepare for the exams if you want to get certified.
We have tried to engineer these demos to be run on a single laptop/desktop running Hyper-V either inside Windows Server 2012 or on Windows 8. The exceptions are where the demos are showing advanced features like the virtual machine migration options, replica, and Hyper-V.
To get you started particularly if you’re new to Hyper-V we need to introduce you to how to build a lab setup. Actually there’s some good resources on TechNet for this but I also wanted to show what to do to get to that, because these guides assume you have your virtualisation setup and you know how to build virtual machines and configure them.
Build your Virtual Machine Host for demos using Boot to VHD
Rather than fiddling around with partitions Windows 7/2008 R2 and later allow you to boot from a Virtual Hard Disk (VHD) rather than a real disk you can have multiple VHD each with its own operating system and each will have a corresponding entry in the BCD on your system. Each of these can be copied around and restored if things go wrong. So lets get started..
Download Windows Server 2012 and start to install it. To do this you might want to use the Windows 7 ISO to USB utility to make a bootable USB stick form this iso file. Start the installation and as soon as you have a dialog box up in the Pre execution Environment (PRE) stop! Hit SHIFT-F10 and this will bring up a command line.
list volume this will help you identify which drive you want to use to host your VHD e.g. Drive D:
create vdisk file=”D:\SysPrep.VHD” type=flexible maximum=20000
this create a VHD, sysprep.vhd (you can call yours whatever you want) that is dynamic and 20Gb in size
select vdisk file=”D:\SysPrep.VHD”
Now you can go back to the installation environment and customise the installation to install the operating system to your new volume.
When you have completed the installation you will want to find any drivers you need to get the display working properly as well as your various network card. I find that the windows x64 drivers are generally Ok for this if devices aren’t detected automatically. You may also wish to deploy the Remote Server Administration feature on your new deployment so you have the tools to manage all the new features in Windows Server 2012 as well as the Hyper-V role (which needs a reboot).
Once you have your new installation the way you like it sysprep the virtual machine (c:\windows\system32\sysprep\sysprep.exe) and set it for an OOB experience and to shutdown (not restart).
Boot the machine from the installation media again and again enter SHIFT F10 to get a command prompt.
create vdisk file= “D:\Boot.VHD” parent=”D:\SysPrep.VHD”
this creates a differencing VHD, boot.VHD with a parent of the sysprep.vhd you created earlier
select vdisk file=”D:\Boot.VHD”
list volume this will help you identify which drive your new VHD has been mounted to e.g. drive V:
this creates a new boot entry that will boot from the difference disk you have made
Reboot the machine and select the top boot option. The machine will come out of sysprep and all of the changes this makes will be written into the differencing disk leaving the parent disk unchanged (in a pre sysprep state). Go into the system configuration of you machine and remove the second boot option (the one that points to the sysprepped VHD that we started with) to ensure you don’t ever boot into that.
With setup you can use the sysprep.VHD as a parent for your VM’s and if you do one more thing..
copy Boot.VHD Boot-Backup.VHD
You can get back to a sysprepped state by copying the boot-backup.vhd over Boot.VHD to get back to where you started from. You could also back these files up to an external drive and copy them in again or onto another machine to repeat the process. Also I have a quick introduction to other BCD related command here which may be of interest.
Anyway in subsequent posts I’ll go through what you can do now you have this setup.
In my last video I showed you how to use Server Manager to manage lots of servers, and given that you want to do that, there is less of a need to have all the management tools on every server. So in this screencast I wanted to show you how to rip parts of the interface out of Windows Server to create a minimal UI (known unofficially as MinShell).
Server Core is stall an install option in Windows Server 2012 but you can now add in the full or partial UI post install. However to do that you’ll need access to install media for windows services (specifically the sources\sxs folder) as the binaries for the UI won’t have been copied as part of the core install process. Because MinShell is simply a feature removal rather than an installation option. You can enable/disable the full use interface whenever you want to. Patching and the reduced attack surface of MinShell are the key benefits for doing this, while another benefit for servers that aren’t in data centres is that casual users won’t be able to go surfing on them as internet explorer won’t be there, as will other tools like explorer.
Server Core is stall an install option in Windows Server 2012 but you can now add in the full or partial UI post install. However to do that you’ll need access to install media for windows services (specifically the sources\sxs folder) as the binaries for the UI won’t have been copied as part of the core install process.
Because MinShell is simply a feature removal rather than an installation option. You can enable/disable the full use interface whenever you want to.
Patching and the reduced attack surface of MinShell are the key benefits for doing this, while another benefit for servers that aren’t in data centres is that casual users won’t be able to go surfing on them as internet explorer won’t be there, as will other tools like explorer.
If you want to try this you just need to download an evaluation copy of Windows Server 2012
My favourite feature in Windows Server 2012 is it’s ability to manage and be managed. For those of you that aren’t yet PowerShell fans, then this means Server Manager and that's the main thing I am using in this screencast..
In a small business you could get away with just using server manager and PowerShell to manage your servers, and in my opinion you’ll be fine where the number of server you have (physical plus virtual) is less than about a hundred). However when you get close to that you need to start thinking about dedicated management tools like System Center 2012
You can manage older servers (back to Windows Server 2008 & 2008 R2), but you’ll need to pull down the Windows Management Framework 3.0, and then run winrm quickconfig if you aren’t already remotely managing them.
You might want to run all these tools from Windows 8, rather than connecting to a server via remote desktop. You’ll want to download the Remote Server Administrator Tools (RSAT). before you ask, RSAT for Windows Server 2012 can only be installed to Windows 8 , in the same way that RSAT for 2008R2 only works form Windows 7.
Hopefully you will have read the introduction to our Evaluate This series and are now ready to start to have a look at how stuff works in Windows Server 2102. The obvious place to start is Hyper-V as in subsequent videos in the series well need a number of virtual machine, and not everyone knows how to do this in Hyper-V. For example some of you might be new to Hyper-V because you are a DBA, A Vmware expert, for example so hopefully this video will help..
It did occur to me that you may want to try Hyper-V in Windows 8 as well and I didn’t cover that off in this video so from the start menu type programs and look for Programs & Features in Settings..
tand then select Turn Windows features on or off from there ..
and select Hyper-V..
You’ll need to reboot, make sure your BIOS is setup to support virtualisation and even then this may not work if your CPU doesn’t support SLAT (intel = EPT, AMD=NPT) and you can test for this use CoreInfo (part of Windows SysInternals ) if you’re not sure.
1. This is just an introduction to Hyper-v to help you setup your first basic virtual machine as I want to keep each video as short as possible and there will be others in the series to show you some of these other advanced features.
2. In the video I use a sysprepped copy of Windows Server 2012 as a parent disk to create new difference disk to use in my virtual machine, and so please refer to the introduction to this series for details on how to make that. If you haven’t got a TechNet subscription you can use the Windows Server 2012 evaluation edition
3. If you do want to try this in Windows Server 2012 and you haven’t got an MSDN/TechNet subscription there’s a 180 day Evaluation Edition of Windows Server 2012.