I'm using Windows 8 Pro RTM and am attempting to configure an External Virtual Switch.  I have an Asus P8Z68-V and it has a single Intel 82579V controller.  I'm attempting to use the option to share the NIC with the management OS.  When I attempt to apply the changes, I get the following error:

[Window Title]
Virtual Switch Manager

[Main Instruction]
Error applying Virtual Switch Properties changes

[Content]
Adding ports to the switch 'External Virtual Switch' failed.

The operation failed because the file was not found.

[Close]

I also have this in the System event log:

Once this error message occurs, the NIC has a Device code 10 (cannot start) and network connectivity is lost for the host.  I have to scan for hardware changes and the NIC comes back online. Thoughts on what could be wrong?

http://sharepoint.nauplius.net

I'm using Windows Server 2008 R2 with Hyper-V installed.  It is a standalone machine.  I have set up an internal virtual network but only for some of the VMs.  When I try to start VM not in the internal virtual network that has no network adapter, I will get the error: Microsoft Emulated IDE Controller: Failed to Power on with Error "The process cannot access the file because it is being used by another process."   If I turned off ALL VMs in the internal virtual network, I can start that VM.  Then, if I try to start a VM in the internal virtual network that is using network adapter, I can start it.  However, if I try to start a VM in the internal virtual network that is using legacy network adapter, I get the same error.

When running Edit Virtual Hard Disk Wizard, it hangs up on Getting disk information I've been sitting on this screen for 5 hours now.

the server is in the off state I do not see any snapshots and I ran DISKSHADOW>Delete Shadows All just in case.

How do I get past getting disk information screen?

I wanted to move a VM from an older Windows Server to a new machine running Windows 8 (actually they were the same physical box - I just replaced the old OS with the new one). As the new Hyper-V should be able to just open VMs without the need to export / import them, I only backed up the files and restored them under Windows 8. Only then I found out that the client-side Hyper-V is actually stricter in system requirements than the server version and unfortunately, my CPU doesn't support SLAT.

So what I'm trying to do right now is to make the VM run under VMware Workstation and it crashes very soon after I see the "loading Windows" animation. There may be all sorts of reasons for this but before I investigate further, I remembered that I am actually not sure if I turned of the VM properly. It was in the "Off" state when I made the backup but I'm not sure if it was shut down or just suspended.

Looking at the files, I see there are .vsv and .bin files under the VM root and those are probably memory snapshots. Are they always there or do they indicate that I have just mistakenly suspended the machine and not shut it down properly? If that's the case, will I be ever able to open such VM, even if I manage to get Windows 8 Hyper-V working?

I backed up my VMs by exporting them, installed Server 2012 and re-imported them. I am having issues RDP'ing to VMs that are on an internal network. The host is configured as a router.

If I change the IP and network to the physical NIC I can connect OK.

The error I recieve is "The Kerberos client received a KRB_AP_ERR_MODIFIED error from the server server$. The target name used was ldap/server.name. This indicates that the target server failed to decrypt the ticket provided by the client. This can occur when the target server principal name (SPN) is registered on an account other than the account the target service is using. Please ensure that the target SPN is registered on, and only registered on, the account used by the server. This error can also happen when the target service is using a different password for the target service account than what the Kerberos Key Distribution Center (KDC) has for the target service account. Please ensure that the service on the server and the KDC are both updated to use the current password. If the server name is not fully qualified, and the target domain (domain.name) is different from the client domain (domain.name), check if there are identically named server accounts in these two domains, or use the fully-qualified name to identify the server.

Hello,

I came to seeif you have afeedback from theIntelteaming.Here is thedescriptionofmy lab(W2K8R2SP1):

-Hyper-V2- 64GB - 14Broadcom NICports - 1NICIntel (integrated- Modeli350) 2ports

- Broadcom: 2 portswith 802.3adCSV(10.0.0.x)- 2 portsiSCSISLB (192.168.1.x)

- Intel: 2 portswith 802.3adProduction(Subnetting:172.10.0.x =>Hyper-V -172.10.10.x=>VMs)

-Network SwitchHP Procurve1910 -ports configured asLACP- disableSTP- VLAN...

I'm having problemsreally strange, andI think I haveisolatedthe problem:Intel NIC Teaming.I have noproblem with myBroadcom adapter,bycons withIntels,I'm going crazy,theproblemcompletelyrandom andpingsare involved:

-Sometimes Icould pingany server, sometimes not

-I canpingan hyper-v from VMs, sometimesnot

-From avmimposibleto pinganothervm

-From avm, I canpinghyper-v,sometimes not...

-In anycase,whenI run aping, I have afirst responseafternothing ...

-In anycase,whenI run aping,I have noname resolution

-In anycase,whenI run aping,I don't haveanswers right away, but after2/3 minutes, I getanswers ...

In short,totally random problem, but when I disabled LACP on theswitch,it seems to be ok...(but not very sure)

I know the policy supportof MS (http://support.microsoft.com/kb/968703) aboutTeaming,integrated cardsthatare not recommended,asthis is a known problem, etc ...

Even more bizarre that on another server, I had severals VMSMP errors (Port '97B1A2BA-D679-4290-AD9B-262DE33C16D6 'was Prevented from using MAC address '00-15-D5-04-0C-F8' Because it is pinned to port 'BFA4D592-1F9D-8484-8ABF-9EDA1534529E'.) and I solved this by changing the type of Teaming (Intel NIC too)!

Ijust looking for afeedback,whetherI pushfurtheror not, orif I buy aBroadcom;-)

-- Cédric GEORGEOT [MVP] Virtual Machine http://www.e-novatic.fr -- Auteur du livre "Bonnes pratiques, planification et dimensionnement des infrastructures de stockage et de serveur en environnement virtuel" -- N'oubliez pas de marquer comme réponse ;-)

Hi All,

An easy score for someone here I think but I can't get a definitive answer to this through my searches.

I am in the middle of some Hyper-V trouble-shooting and have decided to dedicate a NIC for CSV traffic, to the point I am wondering if the CSV NIC needs to have the - allow cluster communication - option enabled. I have this enabled on both my live/heartbeat and management networks and wondered if I can lose it on the CSV NIC to really narrow down my findings.

Many thanks

I hope someone can help me with this. Here is the scenario I am faced with.

I had VMs running on an external ESATA raid array (E:) off one of my test servers. Unfortunately it had a failure and had to move the VMs to a new external ESATA raid (G:).

I noticed that even after the VMs where exported from the old raid and then imported on the new raid, the VMs are still mapped to the old raid. I can't turn off the old raid due to this reliance. I have tried to remap all the VMs accordingly, but there is still something keeping the VMs mapped to the old configuration files. I cannot find anything that is pointing to the old raid. All VMs with this reliance go critical when the old raid is turned off. This is all done with Hyper-V manager.

Joseph Yedid

Hi

I am noticing performance issues on my Hyper-V virtual servers, especially with my terminal servers, Exchange box and SQL.

Setup on the 2 Physical servers themselves are Windows 2008R2 Enterprise, over 60GIG RAM on each node, both nodes are in a cluster, storage is on a iSCSI SAN and the Hyper-V traffic is run through two of the four ports on a NC375T HP NIC teamed with HP Teaming.  The VM's have fixed HDD's, I can't see any issues with performance around CPU or RAM usage.

I have been reading alot about disabling IP4 Checksum Offload or offloading etc on the NIC's, not sure on the phyiscal NIC themselves, the Microsoft Virtual Machine Bus Network Adapter on the VM's or under the Teamed Settings on the physical hosts...

Any idea's or hints would be great.

Thanks
Mark

All, I am in the process of creating my first VH to host 3 VMs.  I have 3 - 300GB drives that I need to configure to be the host and VMs. This is what I was planning on doing.

1. Create two arrays with a hot spare.

    a. 1 array for the HOST at approx 120GB

    b. 1 array for the VHDs and Config files, each VHD will be about 120GB.

2. I then have an attached storage array the will be used as extra storage for each VM that will be configured at a later date.

Does that sound like a logical plan.

Hi,

I am exploring the options on building an HyperV3 Enviornment with HVR.

Here is my plan.

# 5 Node Failover Cluster on Primary Site

# 3 Node Failover Cluster on DR Site

# Partial set of VMs from Primary Node needs to have a replica in DR Site

I have built the 3 Node Windows 2012 cluster in DR Site, though I havent considered the requirement of replication at that time.

Storage allocation

I have planned for 5 TB in primary site and 3 TB in DR site

5 TB as a single volume or 1 TB x 5 Volume?

Is there any recomandation with respect to thin provisioning?

Thanks

Shaba

Shaba

We have Windows 7-32 running virtually on a 2008R2 HyperV server. Remote access worked fine until I applied the Aug 14th patches and Symantec SEP last week. I can access it from the console, but remoting into it over the network either connects you and then throws you off or you just get an error when you run remote desk top connection. I uninstalled SEP and then went back to a week before the patches were installed. The same issue still occurs. All the services are running though they are set to MANUAL...I tried restting to Automatic but that didnt help either.

Any ideas would be greatly appreaciated. Thanks.

Steve Erlitz

Hi all,

I have two hyper-v hosts diskless boot from a LUN which is located on my NAS storage, and I put these two hyper-v hosts in a failover cluster.  I installed some virtual machines on these two hosts, these virtual machines VHD files are always stored on another LUN, which is also located on my NAS storage.

I think these can enhance the performance of my cluster, because hyper-v hosts and VMs are both on my NAS storage.

Unfortunately the results is out of my expectation. In the two test below, I use gigabit switch to connect hyper-v hosts and NAS. All NICs used is also gigabit. I use the share folder named 'public' on NAS to simulate the virtual machine which is located on NAS.

1. Before I implement diskless boot of hyper-v host from NAS, my two hyper-v hosts boot from local hard disk. I perform file transfer test between two hyper-v hosts and 'Public' share folder on NAS. The read speed on NAS is around 220MB/s, write speed around 120MB/s

2. But after I implement diskless boot of hyper-v host from NAS, the same test give a read speed of only 85MB/s and write speed about 45MB/s

This shows the performance greatly drops after I implement diskless boot from NAS. Am I wrong ? I wish to hear from microsoft expert.

Thank you very much.

At the moment we are planning a conversion project from XEN to hyper-v (Server 2008 R2). There are multiple XEN Servers connected to a fiber channel san. The plan is to set up two new Servers with hyper-v move all the virtual machines to the hyper-v servers, delete the san, make new luns for quorum and  the cluster shared volume, make a cluster from the hyper-v servers and move the virtual machines to the csv.

My question is now if this way would work or if we should better set up the cluster before adding virtual machines to the hyper-v hosts. The alternative would be to set up the cluster on a iscsi nas and move the quorum and csv partition to the san later.

It would be great if you have some suggestions for us.

I am exploring the options on building an HyperV3 Enviornment with HVR.

Here is my plan.

# 5 Node Failover Cluster on Primary Site

# 3 Node Failover Cluster on DR Site

# Partial set of VMs from Primary Node needs to have a replica in DR Site

I have built the 3 Node Hyperv3 Windows 2012 cluster in DR Site, though I havent considered the requirement of replication at that time.

Please give your recommandation which can fit my requirement

How many network interfaces

I have planned for

3 NIC for HyperV Data - Dedicated VLAN

1 NIC for Management - Dedicated VLAN

1 NIC for LiveMigration/CSV - Dedicated priviate VLAN

1 NIC for Heartbeat - Dedicated Private VLAN

Do we need dedicated NIC for replication?

Do we need dedicated NIC for CSV?

For failover to happen, do we need a streched VLAN?

Can we use F5 GTM loadbalancer to redirect the traffic between a VM on primary site and its DR VM (Once we move to DR mode) ?

Thanks

Shaba

Shaba

Running WINSERVER2k12 RC Build 8400

General question I can't find an answer to:

When Import a VM via Hyper-V Manager GUI, after the VM folder is located and the VM is selected, at step 3 I am asked to "Choose Import Type" and I have three options to choose from:
•Register the VM in-place
•Restore the VM
•Copy the VM

Can anyone please explain to me the difference between  the options?

Thanks!

Vincent Sprague

Hi,

We are setting up a hosted cloud platform and are making use of Hyper-V + Failover Cluster and some in-house developed powershell/php scripts to automate the ability for customers to purchase and setup virtual hosted servers, we have the roll out process and everything working and can provision systems for clients

The problem we are running into is that some clients want physical console access to the machine (ie what you would get using VMConnect) in cases where they want a machine that isnt connected to the public internet, or if they accidentially make changes to the VM that break its connectivity.

We have thrown together using bits-and-pieces a web page using the active-x RDP applet and some automation code that will allow then to remotely access the guest console using the same protocol / ports that vmconnect uses (similar to the self-service portal for SCVMM), the down side to this is:
 - It uses the notsafeforscripting version of the active-x so security settings need to be modified in IE to make it work
 - Registry modifications need to be made to allow the passing of credentials properly
 - It uses the vista and newer version of the control so it wont work on XP
 - Because it uses active-x it wont work for customers trying to connect using Mac or Linux or non-IE browsers

We looked into SCVMM but it again uses the same activex plugins to do its work so it wont work on Mac / Linux.

What we want is to provide some sort of wrapper / glue / application that will take the RDP protocol that hyper-v uses to connect to the guest consoles and convert that into something else (ie VNC, Java Applet, etc) that is more widely usable, we have looked into other products but the only thing even remotely doing development on this side of things is FreeRDP but they only appear to be poking at it and no serious development. Additionally it only appears to work on Linux as it relies directly on X11

Are there any programs / applications / anything that will do something like this??

The only other solution we could come up with is using remote desktop services to provide customers a temoporary login to a remote desktop session server that is locked down and only allows access to the VMConnect applet to connect them to their VM console, but between licensing and trying to automate that it would be a nightmare. 

The only other thing would be to ditch hyper-v and switch to some open source virtualization / cloud management system, but we already have a large install base of hyper-v servers and use the automation services we've developed to manage them, it would be nice to keep using hyper-v for our cloud based offerings

hi all,

I am using windows server 2008 r2 in my physical machine.

i have installed hyper-v role in this machine and have created internal & external virtual network using hyper-v virtual network manager.

i have a VM in Hyper-v which is having DHCP,DNS & AD in it.Now in my physical machine's network connection there are 3 network adapters listed in total.

one Physical NIC ,one hyper-v external network and one hyper-v internal network.

Machine in hyper-v internal network communicates well,hyper-v external network also works fine.

Now my problem is that when ever i enable my hyper-v internal network in the physical machine network connection,Internet in the physical machine goes down so,i have to keep it disabled & because of which my physical machine cannot connect to the internet network.

Transfer of data from physical machine to the internal machine's is a big problem now.Every time i have to create an ISO file to do this.

Please help.

Thanks

Pranay.