I'm administering a 2008 Hyper-v server I didn't build.  It was a 4-port gigabit NIC card.  From what I understand, Windows 2008 server does not support NIC teaming.

I assumed that if I plugged in more than one port the new nic is added to the virtual switch and load balancing is done.  And, when I did this I found 2 entries in DNS with the Hyper-V server's name on them.

Is this the correct way to do this?  I'd like to use all 4 ports and have all of them available to pass traffic to the 3 hosted VMs, but couldn't figure out if all the nics are assigned to the virtual switch or verify load balancing is occurring.

If this isn't possible I'd like to at least assign one nice to each VM and one to the Hyper-V server.

Looking for ways in which I can limit the amount of data any one VM can transfer in a month to external networks in Hyper-V 2012. Ultimately something which tracks data consumption live, and also has the ability to send warning alerts as usage limits near and finally disable all external traffic once a threshold is reached.

Anyone doing this via scripting, or are people using 3rd party options for this?

I run Windows Server 2012 RTM Hyper-V and I can occasionally not access the second virtual hard drive (dynamically expanding VHDX) attached to the VM through the virtual SCSI controller. I can however access the first hard drive that is connected with the virtual IDE controller.

I get the following warning in the event log under “Administrative Events” every 30 seconds when this happens:

• Log Name: System
• Source: Storvsc
• ID: Storvsc
• Message: “Reset to device, \Device\RaidPort0, was issued.”

I get this error once or twice a week and it has caused serious problems since one of the virtual servers that have this problem is a fileserver and the second hard drive contains all the data.

The only quick solution to the problem that I have found is to force the virtual machine to stop using the “Turn Off” feature since a normal shut down does not work (stops at shutting down the event log or similar) and then start the virtual machine again.

You can also wait for about 30 minutes or longer until the disk for some reason becomes accessible again by itself.

My research into this problem shows that:

• Only 2 of the 10 VMs running Windows Server 2012 RTM that I have, have this problem.
• Both these VMs have a second virtual hard drive (dynamically expanding VHDX) that cannot be accessed for 30 minutes or longer.
• Check Disk of the virtual hard drive shows no errors.
• The second hard drive is attached to the virtual SCSI controller.
• I can find no problems at all with the physical storage on the (not related) 4 hosts that I have. The problem exists only in the VMs.

I have now attached the second virtual hard drive to the virtual IDE controller to see if this permanently fixes this problem (i.e. does not happen for at least a week).

Is there something wrong with the virtual SCSI controller or the virtual SCSI device driver that comes with Windows Server 2012 RTM? Does anyone else have this problem?

Hi,

We have about 70-80 different offices at different locations. Each location has a HyperV server and 3-4 virtual machines on it. We are not using HyperV Core - It is Enterprise Windows 2008 R2 - Full installation with HyperV Role.

We need a backup solution which can backup all the data of the host and VM's onto a Tape Library. Many backup solutions (e.g EMC Networker) support HyperV backup however require a separate server. The support matrix does not include Windows 2008 with Hyper as the Storage Server. We cannot buy another server for backup due to cost.

Any suggestions on some good, cheap backup solution which we can install on HyperV Server?

We dont intend to use Microsoft DPM.

regards,
Vik

Regards, Vik Singh "If this thread answered your question, please click on "Mark as Answer"

The issue has happened on 2 different hosts in the last two weeks, and we haven't been able to find out exactly what is causing this so I am hoping to get some help.

5 Node Server 2012 Hyper-V cluster.  Monitored using SCOM and also using VMM (2012 SP1).  In VMM we noticed that the server in question was showing as not responding, all the VM's were fine, and VMM shows that the WinRM service is not OK.  The service is running on the host, but the host cannot be RDP'd to, we can't login to the console, and we can't connect to the cluster via Cluster Manager because in both instances the host in question was the Cluster owner.  Our "solution" has been to shutdown the VMs on the host and then forcefully reset it, and after that everything is fine.  However, I am hoping to get to the bottom of what is going on.

This is the exact same issue as described HERE, but there isn't a solution given.  Here is a chronology of events that I gathered from the event logs.

From the logs you can tell that the issues started on 7/7/2013 at 5:38 PM

From Application Event Log:

Error 26001, Microsoft.SystemCenter.VirtualMachineManager.2012.Report.VPortUsageCollection

Got null results from Select Connection from Msvm_SyntheticEthernetPortSettingData where InstanceId='Microsoft:3E323714-F9A3-4384-A2D7-3466B3FED595\\6D14D559-7676-4D9B-82A7-F5F1199601FA'  .  Different instance IDs.

Basically happened every 30 minutes until 1:49AM on 7/9.  I rebooted the server around 11:45 PM that night.

From System Event Log:

Event 1, VDS Basic Provider

Unexpected failure. Error code: 48F@01000003

There are a lot of these errors.  These errors still occur even when the WMI Performance Adapter messages are appearing normally

7/7 5:26 PM

Event 7036, Service Control Manager

The WMI Performance Adapter service entered the stopped state.

7/7 6:06 PM

Event 7036, Service Control Manager

The WMI Performance Adapter service entered the running state.

7/9/2013 8:27 AM

Event 7000, Service Control Manager

The Device Setup Manager service failed to start due to the following error. The service did not respond to the start or control request in a timely fashion.

A lot of these errors

7/9/2013 8:35 AM

Event 7011, Service Control Manager

A timeout (30000 milliseconds) was reached while waiting for a transaction response from the SCVMMAgent service.

7/9/2013 8:46 AM

Event 7001, Service Control Manager

 The System Center Virtual Machine Manager Agent service depends on the Windows Remote Management (WS-Management) service which failed to start because of the following error:

The service has not been started.

Once the WMI Performance Adapter Service stopped stopping/starting again, in the Operations Manager log you start seeing warnings like these:

7/7/2013 5:30 PM

Event 21402, Health Service Modules

• Module was unable to connect to namespace 'ROOT\MSCLUSTER'

• Module was unable to connect to namespace 'ROOT\CIMV2'

• Summary: 1 rule(s)/monitor(s) failed and got unloaded, 1 of them reached the failure limit that prevents automatic reload. Management group "MY_DOMAIN". This is summary only event, please see other events with descriptions of unloaded rule(s)/monitor(s).

• Forced to terminate the following process started at 5:27:36 PM because it ran past the configured timeout 180 seconds.
• Command executed:        "C:\Windows\system32\cscript.exe" /nologo "ConsecutiveSamplesTwoThresholds.vbs"A_SERVER_NAME

Followed by errors like this:

7/7/2013 5:37 PM

Even 22402, Health Service Modules

Forced to terminate the following PowerShell script because it ran past the configured timeout 30 seconds.

 Script Name:        PowerShellScript

One or more workflows were affected by this. 

Workflow name: Microsoft.Windows.HyperV.2012.VMReplicationHealth33412.Monitor

Instance name: A_SERVER_NAME

Instance ID: {EA9D5CBC-577D-C262-CBFC

Forced to terminate the following PowerShell script because it ran past the configured timeout 30 seconds.

 Script Name:        PowerShellScript

One or more workflows were affected by this. 

Workflow name: Microsoft.Windows.HyperV.2012.VMReplicationHealth33414.Monitor

Instance name: A_SERVER_NAME

Instance ID: {EA9D5CBC-577D-C262-CBFC-4F5037B38E50}

Management group: MY_MANAGEMENT_GROUP

The WinRM Service was running on the host (I was able to check it remotely using Powershell), I cannot say for sure about the SCVMM Agent Service, I don't remember.

MCITP | VCP4

Hi,

I have the following question regarding the network configuration in Hyper-V:

The Server has 2 dualport 1GBit/s NIC's with SR-IOV functionality. Both ports of each NIC are connected to 2 different switches. These 2 switches connects a second datacenter (2 Gbit/s interlink) , where another server with same config is located.

From these 2 servers I want to build a cluster for VM.

When I follow best practises for Hyper-V, I should supply different networks: Management, LAN, Cluster, Live Migration (Storage is SAN based). So I could use each of the ports in the dualport NIC's for the different networks and would have SR-IOV support - correct?

Now I'd like to get this more improved (fail-safe) and want to use NIC teaming (one team with NIC port connected to each switch) - but what I have learned is if using nic teaming SR-IOV is not possible, correct?

So what would be the best configuration in this case when SR-IOV is a concern? (Unfortunately the network infrastructure with its bandwith limitations is predefined and not that easy to change)

Any ideas and "real world solutions" appreciated :-)

Regards,
Markus

Hey,

I have 2 Windows Server 2012 Datacenter Edition servers running Hyper-V on them. On each host I have a server that I am trying to setup guest clustering for a fileserver and they both are sharing the drive via iSCSI. But when I try to create the cluster it fails on the forming cluster step. However when I have the nodes on the same Hyper-V host I am able to successfully create the cluster. When I had it working on one host I migrated the VM to the second host and the cluster service on that server failed. Any ideas on what it could be?

Hello Gyus,

I have very big Proplem and I really need your support guys, I have Hyper-V cluster with four node hosts, all hosts running windows server 2008 R2 SP1 with all Hyper-V hot fixes installed, I have created 1000 VMs using VMM 2012 Console successfully, the environment is to be used with VDI, VMs has assigned Legacy Network card to boot from network and get Windows 7 OS, everything is working perfectly, however, if I restart one host, the VMs will be moved to other hosts and stuck on Pending for ever, I have then to disable autostart from the cluster console and start the VMs one by one.

Please let me know if there is a solution for this issue, it is illing time and if for some reason one host failed while production, people wont be able to work.

Regards..

Nour

Nour

Hi,

We are planning to have 2 server rooms. In each room we gonna have 100 VM's. Now if I create failover cluster can I replicate it to get even more security because if we lose SAN so everything will be gone. Can I replicate failover cluster or to be more specific SAN so if something happens with first site that VM's can be online in a moment (can I point to another SAN directly without restoring VM to a new SAN or) in the other site?

I´m having this "disk bus type does not support clustering" thing when evaluating stroage in FCM. I´m creating a lab, which requires cluster, I´m using Hyper-V 2012 on a HP Powerfull workstation with LSI SAS controller and 2 SAS disks. I´m having 2 W2012 nodes on a same Hyper-V host, which will be members of a cluster.

With the same hardware and VMware, things worked like this:
- I had 1 mirorred array of 2 SAS disk on LSI controller.
- When adding a disk to VM node, I enabled sharing/pass-through option of desired disk
- I created a volume in first node in Windows Server's Disk Manager
- As iscsi, evaluation of a storage was done and everything worked as a charm

With Hyper-V, I don´t know how to do iscsi. Now my disks are as SCSI, and not supported. Question is, how to create a storage correctly, supported my FCM?
- Should I keep array in LSI? Or should I do single disks?
- Should I create vdmx inside the Hyper-V host, or should I connect disks directly to a VM nodes?
- Should I stick with scsi or fiber channel?
- Which is better, to create a mirrored drive in LSI, or with Windows´s disk manager?

We are preparing to replace a 2008R2 Hyper-V cluster with a 2012 Hyper-v cluster.

We were thinking that we would come in on a weekend. shutdown the 2008R2 boxes, move the San connection to the 2012 boxes and configure the cluster. All of the VM files would be on the SAN then we would use import-vm to just import all of them.

We setup a test lab with the 2012 server cluster on a small SAN for testing. We copied a few of the VM directories over to the lab system.

What we have found is that we can use the wizard to import the VMs with no problem and even with snapshots.

But if we use PowerShell import-vm we get errors telling us to run compare-vm. When we run Compare-vm it tells us the problem is incompatible snapshots.

So I guess the question is: Why can the wizard do it and not the underlying powershell?

Any way around this? I really do not want to do them one at a time with the wizard.

Thanks in advance.

Hello,

I was migrating from Hyper-V 2008 R2 to Hyper-V 2012 and ran into some issues.  Rather than export my VMs I just wanted to import the config after I upgraded to Hyper-V 2012.  This seemed a lot easier than export/import requirement from 2008 R2.

Anyway when doing the "Import" I kept running into a whole lot of Access Denied messages.  After finally giving up I recalled that all the VMs config files were on one or several CSVs.  When importing a VM I dont recall it being marked for HA or what host I was trying to import the VM on.

If you attempt to import the VM on a host that does not own the CSV directly is it able to restore the correct permissions to the VM config files and vhd files?  Some of the VMs I was trying to import might have resided on a CSV owned by a different node in the cluster. 

My theory is that without direct ownership of all the CSVs used by the VM the host wont be able to restore permissions to the VM files and bring the VM online.  Even though the non-direct access to the VM files stored on the CSV should be handled by the CSV coordinator/owner.

Running into a random issue, that impacts a couple of our virtual machines.  First our setup:

3 Host Cluster - 2 * Dell PowerEdge R710 and 1 720.  Connected to Equal Logic SAN.  Running in a CSV.  All servers have latest BIOS/Firmware for NICs.  They are all running 2008 R2 Datacenter with SP1, and fully patched as of last week.

Symptom ->  When restarting a VM on a host, it will shut down, and then fail with the error listed in the question.  Looking at Event Logs, it indicates it "Failed to open Attachment" points to C:\ClusterStorage\Volume# and the location of the VM/VHD.  Second part of the error "A Device attached to the system is not functioning".

We have already tried the FSdepends fix on all the Host Servers.  We have also removed all A/V software from the hosts.

The fix I am using right now, is via SCVMM.  I "migrate storage" to another Host, and then migrate the VM.  The VM ALWAYs will start up again - until the next time I have to do a shut down.  Then I have to migrate the storage, and repeat.  I have 3 or 4 VM's that will do this, regardless of what host they are on.  It is the migrate storage that corrects the issue - if I just migrate the VM, it will not start.

I am at a loss - ?

I'm administering a 2008 Hyper-v server I didn't build.  It was a 4-port gigabit NIC card.  From what I understand, Windows 2008 server does not support NIC teaming.

I assumed that if I plugged in more than one port the new nic is added to the virtual switch and load balancing is done.  And, when I did this I found 2 entries in DNS with the Hyper-V server's name on them.

Is this the correct way to do this?  I'd like to use all 4 ports and have all of them available to pass traffic to the 3 hosted VMs, but couldn't figure out if all the nics are assigned to the virtual switch or verify load balancing is occurring.

If this isn't possible I'd like to at least assign one nice to each VM and one to the Hyper-V server.

Hi All

I have configured Windows 8 with Hyper V and configured 4 vm's to run on it ranging from server 2003 to windows xp.

I have created an External Virtual Switch and linked it to the only network interface the workstation has. (Intel(R) Ethernet Connection I217-V) All setting are left at default. The mother board is a Gigabyte GA-Z87-D3HP

On each of the servers I have assigned this Virtual Connection with standard Network Adapter.

Now this is where the mystery begins. Only the first two vm's I start up get external network access regardless of which two I start. If I shut down or restart one another then get's external access. It only effects external access.

The host and all four vm are able to connect and see each other but only two can ping eg. the gateway.

The host has full access to the network and all vm's. I am going to try install a second network card tomorrow and see if this fixes the problem.     

Need some help and clarification in regards to being able to Boot a Windows Server 2012 Hyper-V Host Server from a MicroSD card. I have read allot of blogs and technical journals and articles on this but there does not seem to be a clear answer. Can you set up a MicroSD card to boot a Windows 2012 Hyper-V enabled host server from a MicroSD card inserted into a server? Most of the information I see says that the SD card must be set, from the manufacturer, to be a "non-removable" device. Otherwise the card will show up in the "removable" devices registry.  Is this correct or does it matter on SD cards?  Thanks...

Recently we've set up a two Windows Server 2012 Datacenter as Hyper-V nodes in a cluster.
The nodes have dual Intel E5-2620 CPU's at 2.00GHz for a totall of 24 logical processors.
Each node has 16 DIMMs of 8GB for a totall of 128GByte RAM. Low voltage DDR3 modules.

Both nodes have a dual port QLogic 3242 convergered ethernet adapters.
The ports are set up using NPAR to each have four virtual NICs, of which both last two are set to a max bandwidth of 9Gbit/s.
There are four teams created over the four vNICs. Management, Virtual Switch for Hyper-V, Live Migration, and iSCSI.
All are set up as Failsafe Teams, meaning only one port is used actively.

iSCSI teams are set up to use a Max Ethernet Frame Size of 9614, and pings to the Netapp SAN respond to a 8000 packet size ping.

The nodes are all very fast and responsive. It is the VM's that I find too slow in several ways.

On of the main things I've bee noticing with Hyper-V so far, and we still have a 2008 R2 cluster from which I am migrating to the new 2012 cluster, is the relative slow performance of the VM's. Sometimes they are slow to respond to a RDP connection, slow response with starting programs.

The most odd thing I have witnessed so far is Quick formatting a virtual hard drive inside a Windows Server 2012 Standard VM.
A 40GByte drive (GPT, NTFS with Default sector size) takes about 5 minutes. Never had that problem on 2008 R2.
Both clusters connect to the same SAN using the same NIC configuration.

It sounds to me like a I/O problem, but I can't figure out why.
For example: When I migrate VM's from the 2008 R2 cluster to the new 2012 cluster (manually, since the Failover Cluster wizard cannot handle VM's created in SCVMM 2008 R2), and then Convert-VHD them to another drive to a VHDX file, it's runs at 1 to 1.4Gbps in both ways. Oddly enough the Move-Item back to the Original location does it in half the speed (which I assume is a bug in Move-Item).

This makes me believe the I/O is just fine for the host, so why is it so damn slow in the VM?
Although when I copy a file from the local hard drive of the host to a hard drive inside the VM, from within the VM, it shows a 1 to 1.6Gbps speed. Totally fine.

Is it just some weird bugs that exist in Hyper-V 2012, and will those be fixed in R2 or is it something else that I am missing?

Hi All,

We're planning on deploying a multisite Hyper-v 3 cluster, well when I say multi-site, I mean same site but different server rooms, 1 node in each with a 10Gbps link between the rooms.

We'll use a mirrored storage solution with a node in each room.

Are there any nuances or gotchas with this configuration that anyone can give me a heads up about? I have seen a plethora of information on Server 2008 but not a great detail on Server 2012.

Thanks.