Hi, we have a Hyper V 2012 R2 Core cluster of three servers.. We have separate NICS for iSCSI/MNG/LAN/Cluster traffic etc... All is working really well until we create a vSwitch of teamed NICs. As soon as we do that the A record for the Host is updated with the DHCP assigned IP. We have tried using nvspbind to force the ms_tcpip protocol for the MNG NIC to the top of the binding order, but we still have our A records being overwritten by the DHCP enabled Team...

It isnt easy to see the config of these 'adaptors' with the Core version, and I have also searched to see if I can turn of DNS registration, but the articles I have found are not applicable because the Hyper V Team/Switch cant be accessed in the same way - if you know what I mean...

I am sure I have left out a few critical points which you may need to help me answer this issue, so please feel free to come back with further questions and options.

Thanks very much for any help.

Hi,

I set up Windows Hyper-V Server 2012, and I have it running one VM with Windows Server 2012 Essentials installed. This was working pretty well, and I attached 5 physical disks to the VM for storage. With these 5 disks, I set up a storage pool using Storage Spaces. As I copied data into the pool and emptied single drives, I added more and more to the pool until there was just one left. During this time, I restarted the VM, and everything came back just fine.

During the last copy, the source drive became unavailable. Thinking that this might just be a transient hardware error, I decided to restart the VM. This time, the VM did not come back. It could not start because of a "missing attachment." I didn't look much into it at this point; since it was the only VM, I figured I'd restart the whole system (Hyper-V Server). After the system came back online, I found that I still could not start the VM. I finally went to diagnose the problem, and I found that the physical disks that I had attached to the VM are no longer present. I logged into Hyper-V Server and ran diskpart "list disks" - only two of the 6 disks showed up. The other 4 are in a storage pool, but they have gone completely missing. I know that in a storage pool, disks get a different partition table.

1) Shouldn't these disks still show in diskpart, even if they are in a storage pool? Remotely connecting to the volume service via disk manager has the same result. Curiously, the two available disks are "Disk 2" and "Disk 5" (instead of 0 and 1).
2) How can I reclaim the disks?
3) If the disks are no longer visible by design, how can they be attached to a VM?

Thanks!

I used the following code the check the NIC status, but it can't work on Hyper-V with CentOS 5.3 installed:

    if(ioctl( skfd, SIOCETHTOOL, &ifr ) == -1)
        return IFSTATUS_ERR;
    else
        return ifr.ifr_data.data==1 ? IFSTATUS_UP : IFSTATUS_DOWN;

The error is "EOPNOTSUPP (Operation not supported)"

So could you please tell me how to check it on Hyper-V? and which operations could support? Thanks a lot!

Hi All,

I have an issue with a Server 2012 VM running on a Server 2012 std host.

The VM is a file server and has an additional VHD attached which contains the files to be shared.  This 'files' VHD is located on a SMB share and attached via disk management.

This works fine for the Guest and allows portability of the actual storage with out being tied to the guest but the back up of the guest VM fails.

I am using windows server back up and only trying to back up the guest VM not including the 'files' VHD (this is done as a file level back up of the VHD) I select the host under the hyper V tree option.

Other guests VMs back up correctly.

I get an error which can be tracked back to this error in the hyperv vmms log:-

Hi all,

I just migrated my cluster from 2008 to 2012 using cluster migration. But when i try to test live migration, each vm gives following error:

The Virtual Machine Management Service failed to establish a connection for a Virtual Machine migration with host 'XXXX': The target principal name is incorrect. (0x80090322).

The Virtual Machine Management Service failed to authenticate the connection for a Virtual Machine migration at the source host: The target principal name is incorrect. (0x80090322).

It seems like a SPN problem but i do not know which SPN should be registered. Both nodes have same SPN listed.

Same Vswitch configured as well.

Any idea?

I have been scouring the internet for over a week now, and I cannot figure this one out. Some background... I am retiring an old physical file share server, and I am creating a new 2012 R2 VM on a 2012R2 DataCenter Core Hyper-V host. Creating the VM went without issue. It is configured as follows:

8GB RAM

4 CPU Cores

72GB system drive (C:\)

2TB data drive (D:\)

12GB page file drive (Z:\)

I have about 1.8TB of data to copy over from a physical box, so I am using RoboCopy. I have gotten about 1.5TB of data over, then the new VM's data drive (D:\) disappears and the copy fails. The disk no longer appears in Windows Explorer or diskmgmt.msc. The even log shows many event IDs 153 (The IO operation at logical block address 0x*** for Disk 2 was retried) and event IDs 51 (An error was detected on device \Device\Harddisk2\DR2 during a paging operation, and the last event when the drive disappears is a single Event ID 157 (Disk 2 has been surprise removed).

After racking my head over this for several days, I deleted the VHDX for the 2TB data drive, created a new one, and started the data copy all over... same issue. I completely deleted the VM and all files, reinstalled on new VHDX drives... same issue.

This leads me to think it is an issue with the Hyper-V host, but there are no event IDs pertaining to this, and the other 53VMs are running without issue.

I would SINCERELY appreciate any help with this.

Sorry for huge post, but thank you in advance!

Hi,

we have setup a new server with Windows 2012 R2 and migrated our VM's over to it.

OS within the VM's are Windows 2008 Server, Windows 7 and Windows 2012. 

Everything works fine. Integration Service Update was needed and I could install it on all VM's, except one, that also runs Exchange 2013 on a Windows 2012. 

Once I install the Update of the Integration Service, the VM acts weird. It takes about 30 minutes until I could log in. (Probably network problem after the udpate). Once logged in, if I switch to desktop view, the task bar doesn't show items other than the clock, clicking it results in the question, if I want to end a hanging process no matter if I say yes/no no luck. Server Manager opens (very slow), but it doesn't load data (wanted to check nic).

CMD with ipconfig doesn't show a network adapter, just the tunneling-peudo-interface.

 Load of the server doesn't show high CPU or Memory consumption, nor high disk i/o. Using Alt/TAB it shows a window called "Immersive Background" (tried to translate to english). 

Starting the network manager via

control.exe /name Microsoft.NetworkAndSharingCenter

doesn't open it either. But I can open like Resourcemonitor, Powershell and quite some other programs. 

So something is failing here after the update.  

Does Exchange and the new Integration Service break each other? 

I rolled back VM to an older snapshot and tried the same thing again, same result. On our Replication host, I did a Testfailover, with no network connectivity, everything fine, again, until new Integration service are installed/updated, the host seems broken. 

So I have no clue what's happening here or where to start to search. The main problem is, that backups have to be done offline and not online until new integration services are up and running. 

Any guess what to try or what breaks the new integration services? Or is it really in conjunction with Exchange? 
We didn't upgrade the VM to 2012R2, just the root node. 

Thanks

Patrick

I have a virtualized Exchange 2013 server on Windows Server 2012 R2 with 2 virtual disks (C: OS drive and D: Exchange database and log files). After a power failure, the virtual server will not start because the second VHDX drive (the one with the databases and log files) is corrupt and unreadable. If I remove the second VHDX, it will start successfully but the Exchange DB's are on the corrupt VHDX.

Nothing worked from other posts and some of the stuff will not work for the VHDX that people are posting for VHD's. I have tried mounting it on a host machine but it errors out saying it is corrupt and unreadable. I have tried to convert it to a VHD but it does not work because it is corrupt and unreadable. 

Anyone have any other ideas? I thought the whole idea of the VHDX was to prevent this from happening?!?

Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread.

Hi,

We are using Hyper-V on Windows Server 2012. We have three virtual servers defined under Hyper-V. One of these is our terminal server and it is re-starting every couple of days without any reason. I have gone through the event logs. There is no event recorded 15 mins before the server re-starts. Hyper-V records the following event details :

Host-Server Error : 18590 Critical Microsoft-Windows-Hyper-V-Worker Microsoft-Windows-Hyper-V-Worker-

'Terminal-Server123' has encountered a fatal error.  The guest operating system reported that it failed with the following error codes: ErrorCode0: 0xD1, ErrorCode1: 0xFFFFFFF0, ErrorCode2: 0x2, ErrorCode3: 0x1, ErrorCode4: 0x44D2C22.  If the problem persists, contact Product Support for the guest operating system.  (Virtual machine ID 3A25CCF9-BC65-4422-A69B-A629E719140A)

Can someone push me in the direction of a probable solution ?

I'm running into a bit of a problem here.

I have the following two virtual switches setup on one 2012 R2 Datacenter Physical Host w/ one (1) physical NIC:

Physical Host is running 192.168.50.120, GW: 192.168.50.1, SM: 255.255.255.0 (Cisco Router has the IP of 192.168.50.1). The Internal vSwitch Network (will contain 8 VM Servers) is 192.168.1.1. I want the VM Servers on 192.168.1.1 to be able to get Windows Updates by being able to route through the 192.168.50.1 vSwitch but yet, still want the Networks segregated.

Both of these virtual switches/networks will be running through one NIC.

I initially setup RRAS to use NAT but I'm not sure if this is the correct configuration. It seemed to work but when I went to Activate my Product Key (via MSDN Subscription) it wouldn't recognize the key and therefore I couldn't run windows updates.

However, when I set up two of the VMs on 192.168.1.1 to use DHCP through the 192.168.50.1 Network it would work fine (i.e. DHCP, DNS and GW were all set to 192.168.50.1).

Any ideas on what I might be doing wrong or how I could setup an effective route between the two networks?

Any and all ideas are most welcome and appreciated.

Sincerely,

Wallace

Wallace Davis

I am trying to set or get the "Connection" property from Msvm_SyntheticEthernetPortSettingData through WMI. Even though the operation does not return an error code, I've not been able to acces the above mentioned property. Retrieving or setting other properties of the same class is possible. How am I supposed to access the specified property?

Thank You

Hi Everyone,

I recently created a HyperV R2 cluster hooked into an iSCSI SAN. When I go into the Placement Tab on my Host the options to add a new default path for newly created VM is greyed out.  How do I restore functionality to this function so I may add a new default path?

Thanks,

D

I was upgrading my Hyper V cluster from 2008R2 to 2012R2 and disaster has hit.

History so far

I built my second cluster, with one of a three node cluster and an older server everything went really well up until the point of of "copy cluster roles" from my old cluster to my new cluster.

A couple VMs came over ok but the rest all failed.  I can still have all the VHDs and and the snapshots (checkpoints) so I removed the failed roles and started recreating them all was going fine and then the cluster lost access to the storage although I was able to see it at C:\clusterstorage .

I got the storage back online, it was the virtual switches having dynamic IPs and no DHCP server to hand out an IP ....DOH

However I am still fighting with the other VMs, I have had a few suggestions such as importing the VMs but that that fails. So after alot of trial and error I have used the "New Virtual Machine" in Failover Cluster Manager I am able to get the VMs back this way by pointing to the original VHDs.  I am unable to see any of my snapshots however, I have gone into each VMs settings and changed the checkpoint file location to the snapshot folder, that still does not change anything.  I have read to power down the VMs and they will merge with the snapshots I left them for about 30 mins and nothing happened.

So I now have working VMs but no snap shots (checkpoints) how do I get them back

Hi all

We have 2 HP servers (2012 R2 GEN1 VMs) on the same LAN running RAID 6 on a 10G network.

Live migration works perfectly at speeds of around 350M.

Hyper V replication runs a little slower and keeps crashing around the 600GB limit. We can resync again and eventually the replication finishes. Our VMs are 11TB in size.

Microsoft say it is a network issue, but my question is why should the replication fail when live migration doesn't? We have also set the resync option to happen immediately, but it never does.

We have tried this with compression and without, but the result is the same.

Thanks for your comments!

Scott

On Windows Server 2012 R2 Host / Hyper-V Windows Server 2012 Node - ONLY

We are getting errors with DNS on Hyper V Nodes - Event ID 8015, and NSlookup shows no entry etc.

Hyper V host is fine and connection from the Node to the host is fine - including ping by name.

But DNS connection from the Node to the domain is not working.

On the DNS server we get error  4015.

First time a reboot of the Hyper V host and Node - fixed the issue, but this time we had to do that plus restart the DNS service on the prime DNS server.

Over 20 other servers all fine, DNS is fine - seems to be only Hyper V Nodes.

Both Host and Node patched to latest etc

I've got a choice of running dual Intel Xeon E5420 CPUs (total of 8 cores, 8 threads), DDR2 RAM and No SLAT vs an Intel Xeon E3-1200 Total 4 cores, 4 threads), DDR3 ram and SLAT support to run Windows 2012 R2 with HyperV on. Everything else being equal (Ram, disk I/O, NICs, etc) which is the better platform?I realize with the first CPU, I can effectively assign more virtual CPUs to each VM, but is the newer CPU and memory a better fit.

The VMs in question to be hosted would be a Domain Controller, light traffic web server, System Center essentials without WSUS, and light traffic ftp server.

We have 5 virtual machines that turned off by them selves. We are unable to get them restarted. We keep getting a "Could not find a usable certificate error: 'Element not found' (0x80070490)" error.

This is on a windows Server 2012 machine

Things we have tried:

removed the hyper-v as a role and re-added - no success

re created the certificates - no success

removed and tried to re-import the machines - no success

I see articles about this exact error for WS2008 but nothing for WS 2012.. 

Hi Guys!

Im having some trouble to give permission to an non-administrator user to use theHyper-V Manager. I will explain my enviroment so you can have a better look to help me out to solve this.

Have one Windows Server 2008 R2 ENT SP1 with Hyper-V role installed, in a domain enviroment, and a few VMs created and working. If I connect to the server via RDP with an administrator I have no problem using Hyper-V Manager but if a connect with a non-administrative user I get the old "You do not have the required permission to complete this task. contact the administrator of the authorization policiy for the computer 'localhost'".

I create a Role Assignment with the Read Service Configurationoperation and assign it to the user. I also put this user into the Administrator Role Assignment to discard that I need an extra operation to user de Hyper-V Manager.

Im still getting the error, does anyone have a clue on what Im doing wrong?

Thanks in advance!

We are using System Center DPM 2012 SP1 to backup our Server 2012 Hyper-V Cluster and sometimes we have VM's that are stuck in a state of "backing up..." on the Hyper-V manager however DPM and the VSS Writers both say that the backups have failed. We are unable to migrate or shutdown the guest VM's that are stuck in this state. 

So far the only workaround we have found is to fail all the VM's that are not in this status to another host, then kill the cluster service on the host with the stuck VM's to force them to migrate to another host which also causes an unexpected reboot of the guest. 

I was wondering if there is any other method to cancel or get the VMs out of this state and/or fail them over to another host without having to kill the cluster service?

I have a customer that uses a big line of business app that is written in Visual FoxPro 9 (VFP9).  The application ran very well in a Server 2008 terminal server VM on top of VMware ESXi 5.1 on a Dell T610 with 15K drives in a RAID10 behind a Dell H700 controller (VM was in the RAID10).

We have migrated the customer out of the ESXi environment onto Hyper-V on top of Server 2012 R2.  Net-new terminal server built as a Server 2012 R2 VM on top of new Server 2012 R2 host (Dell T620 with 15K drives in a RAID 10 behind Dell H710p controller).  Host was originally built with a Tiered Storage Space that included all of the 15K drives plus dual enterprise SSD's but I have backed that out.  The application vendor migrated the VFP9 application to the new server and did whatever they needed to do to get the app installed.

The application runs  but it is, to put it mildly, a dog in this environment.  Lookups that took less than a second in the old 2008/ESXi environment can take between 5 and 10 seconds on the new host.  All other applications fly on the new terminal server VM, just the VFP9 app is doggy.  I have tried many things to sort out the problem including running all required VFP maintenance (reindex, repair and pack files) with not much to show for it.  I have rebuilt the underlying RAID10 with differeing stripe sizes, again with no real effect on VFP on way or the other.  I have migrated the VM to other Hyper-V hosts (the T610 has been rebuilt as a 2012 R2 host) and noted similar performance issues.

The app vendor are useless in helping diagnose the problem.  They say the app runs fine under Server 2012 R2 but they only have it running in a physical environment (and I've not actually seen the app run so I have to take them at their word).  I have had to remind them that the original terminal server was also virtualized and did not have issues.

I'd be interested in knowing if anyone else has seen similar behaviour in older apps under 2012 R2 an Hyper-V and, specifically, if anyone has seen this issue with VFP apps?  I am willing to try anything at this point including migrating back to ESXi but I'd rather sort out the issue in the current environment.  Any comments, insights or suggestions would be greatly appreciated.

Robert Dick, Office365 MVP

Robert Dick Office 365 MVP