Evening,

I hope someone can help me with this as I am about out of ideas. Basically I have a vhd that I want to copy/paste into a test enviroment, so I turn the machine off copy the vhd and it fails, I try and move it it fails etc..... all the vhds for this machine move, apart from the c:\ drive. I'll list the things i have tried so far below:

1. Turn off machine, copy then paste into new directory while on the local machine = copies pretty much to the end then errors
2. Turn off machine then reboot, copy then paste into new directory while on the local machine = copies pretty much to the end then errors
3. Turn off machine, copy then paste into new directory on a remote machine = copies pretty much to the end then errors
4. Turn off machine, copy then paste into new directory using a remote machine = copies pretty much to the end then errors
5. Backup with backup exec then restore = backup fails, starts runs for a little while to about 28gb then fails "permission denied"

i tried all of these a number of times and same results each time.

The one time i restarted the machine it restarted to a blue screen about the SOFTWARE registry hive being broken. To resolve this i mounted the VHD on the host machine and tried to replace the C:\WINDOWS\System32\config registry files with the C:\WINDOWS\repair copies. When i tried to copy paste the system file as a backup it not only failed it killed the mount of totally and dropped the drive. It did this repeatedly everytime i tried. It did however let me rename the file and then paste the backup in, which fixed the server.

So right now i appear to be in a postion where I have a VHD I can do nothing with apart from run the machine on it as it appears to be locked in place.

I have checked all the VHD files permissions in comparison to one that does work and all appears to be identicle, I even looked at the hive permissions on the host and all appear correct there as well as all DIR permissions.

I am really lost for what I can even try now to correct this issue as it is a production server I currently dont appear to be able to backup let alone move for any reason.

The host is Server 2008 x64 and the VM is a old school 2003 R2 x64 Standard.

Any help/advice would be greatly appreciated, if anymore information is needed i will gladly supply it if at all possible.

Thanks,

Liam

Hi there
Just checking some details. I have an SBS 2011 machine running as a VM on a HyperV 2008 R2 machine.

Under settings for the SBS VM, I have the following ticked by default.

Are these correct? For Time Synchronisation as I am seeing posts declaring that this should be unchecked for domain controllers.  

Many Thanks

MIS5000

I have clients running around the country with Win 2008 R2 hosts running Win 2008 32 bit guests.  The guests are running our vertical market legacy software.

Last week I set up a test computer with 2 identical hard drives.  On one hard drive I installed 2012 Server with Hyper-v.  The other I installed Win8 with the Hyper-V feature.

I'm getting normal benchmark times matching my 2008 R2 host experience on Win8 Hyper-v, but with Server 2012 Hyper-v my times are 2.5 times slower across the board.

I've tried turning Numa off.  Going from 2 to 1 processors.  Lowering VM ram from 4 to 1 gig with no effect.

Anyone think of any reasons why or other things to try?  I've got clients that want new servers before the end of the year, but at this point I can't deploy 2012 over 2008 till I find out what is causing the performance hit.

I am using an XP image as my test guest.  I'm about to try a 2008 32 bit image to see if that works better.  But I would not think I should see any difference between Hyper-v under Win8Pro vs Hyper-v on Server 2012 for the simplistic purposes of this little test.

Anyone familiar with the differences between the 2 hyper-v's that might explain the disparity?

Hi

I'm trying to understand Saved state backups and exactly what happens in a hyper V saved state backup. Is a saved state backup equivalent to taking a crash consistent backup or is it equivalent to taking a cold backup (offline)?

My understanding since VSS is not involved inside the VM is that these are not application consistent backups. What would a user need to do when restoring a saved state backup? would applications inside the VM have some data loss, inconsistent data? 

Could somebody please explain the difference between a saved state backup and a crash consistent backup. Is a saved state backup considered equivalent to a crash consistent backup?

Thanks in advance!

Hi to all, im trying to acces the windows device manager remotely.

I have found the procedure on technet and on forums, this procedure have no issue when using against 2008R2 server device manager, but when i tried the same for the new windows 2012 it does not work.

I have tried Windows 2012 RTM with full gui as a target and also free hyper-v 2012 server as the target too.

As the source i have tried the 2008R2 MMC and 2012 RTM MMC.

I have Enabled the “Allow remote access to the PnP interface” via domain GPO secondary via local GPO to see if there is some difference.

All the computers are members of the same activedirectory domain which have 2008R2 funciton level (both domain and forest).

All atempts made towards the W2012 platform (ragardless the core or full gui) ends with error:"make sure the computer is on network, has remote managemet enabled and runing Plug and play service and remote registry service. the error was the service is not running"

I have checked all: the remote mangement is configured to enable, and both services are also runing, i have disabled the windows firewall to be sure that it is not blocked anywhere along the way.

I was checking the target server event log to see if there is some active refusal of the remote device mananger request. System and app log does not seem to have any trace. Only in security log I am able to see succesfull audit for the logon (type3) atempt from my source server IP which i think is related to the MMC request.

could you please confirm that this is a bug in the new windows server, or get us some workaround please?

As for the microsft hyper-v server 2012, the remote way is the only posible for device manager, this is a serious issue althou the devman is in read only mode and the sconfig does not provide any other means for accessing the HW info.

I gues that the powershell will be the first workaround in answers, but i would like the remote gui management to work too.

Thank you very much.

Hi Guy's,

I've got a s relativ simple question concerning Hypver-V Snapshots and the replica function.

I've got a 2012HV Cluster  and a Singel Replica Server. The VM's run on the Cluster and get replicated to the Replica Server.

Now i took a Snapshot of a VM running in the Cluster.
As far as i can see this snapshot isnt replicated to the replica server?

Is this correct?

In wich state is the vhd of the replica Server? Is it in the actual "now snapshot" state?
Or is it frozen before the snapshot?

What if i revert the VM to the time i took the snapshoot? Will there be a new replication starting from zero?

Or will everything get inconsistent? :)

Or is the snapshot just hidden on the replica Server?

regards

Stefan

Hi everyone,

I've been testing WS2012 for some time but recently I've been running some tests in the lab with Linux. I'm shocked no one else has reported this error but then I doubt that many people run large Linux VM's on Hyper-V.

If I assign 4 GB memory to a VM it installs and runs CentOS perfectly with very impressive performance. 

However, if I assign more than around 6 GB of Memory (8 GB completely breaks it) the CentOS installer does not even load and shows the following message:

PANIC: Early exception 06 rip 10:ffffffff81c451bb error 0 cr2 0

I would like to add that this also happens on a fully installed VM CentOS 6.3 with IC 3.4. Screenshot attached. Hopefully this can be resolved by an update... :-)

Hello,

I've got a 3 node cluster running Windows Server 2012 on HS22 IBM Blades with 8GbFC HBA's connecting to a XIV SAN.  I am trying to get the virtual HBA's working in Hyper-V Guest running Windows 2008 R2.

I first created 2 virtual SAN Switches one connected to each port on my physical HBA (each physical port connects to a separate fabric).

I add two Virtual HBA's in my Guest, one to each VSAN Switch.

When I go to Zone the HBA's I only see side A of the virtual HBA's not both A and B.  On each fabric I should see two (A and B) WWPN but I only see one.

I can add disk and everything works, however when I go to Live migrate I get the following error due to side B not being zoned.

Here are the Drivers I'm using for my Physical HBA's:

Qlogic 8GbFC HBA - q23wx64W12Storv911025_whql-1507663-07202012

Has anyone see this issue yet?

Hello,

I have Windows 8 RC (not server 2012). I have Enable Hyper-V I have setup an Internal network. This network is available to all the clients and the Host Machine. Next I turned on Internet Connection Sharing on the Windows 8 RC.  After doing so the hosts on the internal network will connect to through the physical adapter on the Windows 8 RC Host as long as their networks are configured with DHCP. If I set a static address on the 192.18.137.0 network (the default for a ICS private network) then I can't make any connections.

I don't mind using the ICS DHCP server except I would like to use a reserved IP address for some of the clients. In particular I want to be able to setup a Domain controller in a VM to go through some training scenarios and DCs and the DNS servers they run need to be at a fixed IP if the other clients on the private network are to find them.

So the question is can you configure DHCP reservations and other DHCP options for and ICN network on Windows 8. This seems to be very important if you want to be able to have internet access from the Hyper-V client machines on a Windows 8 system.

On Windows 2012 you can use Routing and Remote Access which is more flexible but that is not available on Windows 8.

Can ICS DHCP be configured? If so where might I find the reference. If ICS would route traffic for not DHCP configured machines that would work as well. If I could set static network addresses on the client and have them routed through ICS that would be great. However ICS doesn't appear to allow this by default (or maybe at all).

Thanks

Hello,

We're company with around 1900 PCs and we wish to implement Desktop Virtualization in the near future.  We're practically a full Microsoft company, almost all of our servers are running Windows Server 2012 and they are already virtualized.    After reading a lot documentation everywhere, we wish to start a plan in order to run a small pilot first.   I hope you can give us some guidance for some general concerns we have at the moment:

• Is it technically possible to implement a "pure-Microsoft" Desktop Virtualization solution?   (We have EA and SA).       Or, do we really have to include -at some point- the services of third-party providers like Citrix in order to deploy a successful project?      (In other words: why do we need them?).
• The vast majority of our users are only task-workers who only need Office, a browser, a couple of Non-Microsoft applications and interaction with printers, scanners, etc..     So.. what's better in terms of scalability (costs) and manageability:  Pooled Virtual Desktops or Session-based Desktops?
• Is there some kind of tool or calculator available, something that lets you input data like the number of PCs, the type of users, etc.  Something that will help us estimate the kind or hardware, servers or storage that we need?
• Where can we get clear information about licensing in Desktop Virtualization?   We've read a lot about some confusing licensing conditions, like the new CDL, for devices like BYOD, etc.   Is there an updated resource where we can get the whole panorama cleared?

Thank you in advance!

Mc

I recently installed the  Linux integration services v 3.4 on my Cent OS 5.8 virtual machine and I am having issues with getting it to do a proper  shutdown.  The server seems to just issue a shutdown and does not do it gracefully. Is there any other steps i need to do or is this not possible ?  My host is server 2008 r2

Question, can I connect my SCSI tape drive to a physical server and add that as a device on a Hyper-V guest, using Hyper-V in Windows Server 2012?

I know this question has been answered numerous times for Windows Server 2008 R2 and the answer is no, but what about server 2012?

From what I gather VMWare can do it so surely Hyper-V can do it in 2012?

Thanks

I have a 4 year old IBM server running Windows Server 2008 R2. It hosts 4 VM's.

I'm trying to retire this hardware and move the VHD's over to a new Dell server but have a very strange issue with regards to packet loss.

The VM's are as follows:
Windows Server 2008 (not R2)
Windows Server 2008 R2
Windows Server 2008 R2
Debian Linux

I have been shutting the VM's down and copying the VHD's onto the new server, then setting them up again from scratch - rather than using the export/import options. I have never had any problems doing it like this.

When I copy either of the 4 VHD's to the new server and boot them up, everything appears to be okay however we start to see very high packet loss when pinging the VM from any other location in the network. The only exception to this is if I ping from the new Dell Hyper-V server they are hosted on there is no packet loss seen, it's only from another server or desktop PC. There is no packet loss from anywhere in the network to the physical Dell Hyper-V host either.  The packet loss is so high that it makes the servers barely accessible to users. I would estimate it's about 40% loss. A ping looks similar to this:

Reply from 192.168.1.230: bytes=32 time<1ms TTL=128
Request Timed Out.
Request Timed Out.
Reply from 192.168.1.230: bytes=32 time<1ms TTL=128
Reply from 192.168.1.230: bytes=32 time<1ms TTL=128
Reply from 192.168.1.230: bytes=32 time<1ms TTL=128
Request Timed Out.
Reply from 192.168.1.230: bytes=32 time<1ms TTL=128
Reply from 192.168.1.230: bytes=32 time<1ms TTL=128
Request Timed Out.
Request Timed Out.
Reply from 192.168.1.230: bytes=32 time<1ms TTL=128

You get the idea...

We purchased 3 of the IBM servers all at the same time and I have been able to successfully migrate the VM's from the other 2 identical IBM servers to the new Dell hardware, so there is something specific on this server that is causing the problem.

I've updated the Integration tools on the Windows VM's once they are on the new Dell server but it makes no difference. I have no idea where to go from here and we really need to retire this hardware soon.

Does anyone have any ideas? I was thinking I may need to log a support request with Microsoft on this one but thought I'd ask here first.

Thanks in advance.

I  don't know what to do and I am ready to throw in the towel with Hyper-V and Windows 2012.

I have server R910 with 512 GB of RAM, a raid 10 disk array,etc. There is no machine more powerful build by the big iron companies. it has 64 cores,etc.

I created 200 identical virtual machines with one master and 200 differencing disks. The master is 20 GB long and it resides on a Ram Disk, so reading is the fastest. Each VM needs 1 GB of RAM only. The machines are all of them Windows 7. If I select them all in the Hyper-V manager and right-click -> Start, only about 1/4 start. The rest stay like frozen in several stages, but they never finish booting. I guess it is the creation of the swap file, which is 16 GB, and I cannot disable it because no matter what I do, Windows will create a minimal (16GB) swap file anyway. I cannot even redirect this file an external SSD or some similar trick,because Hyper-V does not allow that, like its competitors. As a result of this nightmare, I need to start each virtual machine 90 seconds apart, and for 200 units I need (200 x90)/60= 5 hours in order to fully boot my server. This is a joke.

Does anybody has any idea how to solve this riddle?

ff

Given I have installed a Win2012 ServerCore, I discovered that 2008 R2 Admins will not see it, nor will SC VMM 2012, so drop a VM win 2012 under it (or a 2008 R2 Host) with the Windows interface and yes there it is and I can mange it - BUT. How do I get the Hyper-V and other managemnet tools interaface loaded to manage it when I dont have the Hyper-V role installed on the VM, nor would I as it is a VM not a host. Also SP1 for SC sorts the 2012 server 2012 managemnet but SC SP1 is still Beta - shame. Guess it takes a little time to get all the ducks in a row.

I have always understood servercore is 20% better performing than a Windows GUI bloated VM host so I assume the same is true in 2012. Whilst I am a fan of a CLI the GUI is just easy to navigate and spinning a VM with the GUI tools under it is essential as far as I am concerned, now I just need all the Admin tool sloaded in a Management VM under 2012 - any pack with this all in in I can drop in to manage the network from Win2012?.

Mat

Hi,

I just setup a couple of Windows Server 2012 Hyper-V boxes and enabled the replica of a VM. All seems to work fine. These 2 boxes are connected to corporate LAN through a NIC and each other through a crossover cable over another NIC. As I can see, the replica trafic flows through corporate LAN. I'm trying to move it over the crossover lan, to save corporate bandwith for VMs, but I can't figure out how. I tryed to setup the replica using the IP address of the "crossed lan", but in this case the replica setup fails with a kerberos error (maybe because that lan is detected as public and not as a domain one). I've not tryed with certificate authentication, but this seems to me quite cumbersome considering that both servers are members of the same domain (Kerberos seems the smarter choice).

Has anyone ever tryed this kind of configuration ? What's the best thing to do ?

Thanks a lot.

Massimo Piceni.

I am building a Hyper-V private network to prepare for my Microsoft Exam 70-646.   It needs a 2008R2 domain controller, 2 2008R2 member servers and a W7 client.   I decided to create a virtual network since the I only had one machine at home to build this network.   I want the network to be strictly my 4 virtual machines.   I have my domain controller built and a member server built.   My member server was able to join the domain and talk to my domain controller but my domain controller can't talk to it.   I can't even ping the IP address of my member server from the domain controller.   I'm not sure what I am missing.   

I have completely deleted my network once and recreated it.   I have removed the Hyper-V role and readded it.   I have upgraded my network card in my host machine but I don't think that is it because my virtual NIC isn't pointing to that driver but then again maybe it needs to.   

I do not want my Virtual Network to get to the internet.   I just want them to talk to each other to be able to preform the lab exercises in the training kit.

I'm running Hypser-V on Server 2008 R2 (standard). I have two network cards as per the minimum requirement. The VMs need mostly external communication, although there is some inter-VM communication (between VMs on the same server) that needs to occur. For the inter-VM communication, how much benefit would I get from having an internal network configured?

Thanks,
Nick.