2012 Hyper-V Cluster - Microsoft Hyper-V Network Adapter no connection but Legacy NIC works

July 29, 2013, 2:49 pm

≫ Next: Move-VM - 'Incorrect function.'('0x80070001').

≪ Previous: Server 2008 R2 Ent showing as not all RAM usable

We have a 2 node Hyper-V 2012 cluster on 2012 Standard Server Core.

Cluster Aware Updating was used to update the nodes to current.

4 NIC ports total on each node.

2 connected to management network using standard load balancing.

2 connected to production exclusively using HyperVPort load balancing.

A newly stood up Windows 7 VM does not receive any network connectivity using the Microsoft Hyper-V Network Adapter. Integration Services updated and a reboot and still no connectivity.

Shut down the VM and install the Legacy Adapter and boot the VM up and connectivity works.

How do I figure out where the problem is please?

Philip Elder

MPECS Inc.

Philip Elder SBS MVP Blog: http://blog.mpecsinc.ca

↧

Move-VM - 'Incorrect function.'('0x80070001').

July 27, 2013, 3:15 am

≫ Next: a disk read error occurred - after moving VM storage

≪ Previous: 2012 Hyper-V Cluster - Microsoft Hyper-V Network Adapter no connection but Legacy NIC works

I am trying to live migrate a Hyper-V Guest between 2 Server 2012 Machines using the following powershell command:-

Move-VM vm server Hyper-v Host 2 –IncludeStorage –DestinationStoragePath f:\temp

(names have been changes)

but after the first 2 vhdx files have been copied it fails each time (tried 3 times) with the following error:

Move-VM : Virtual machine migration operation for 'VM Server' failed at migration source 'Hyper-V Host 1'. (Virtual machine ID
99E5A55E-B653-4F1E-9E50-77F96A90F37F) Migration did not succeed. Mirror operation failed for the source VHD file
'F:\hyper-v\vhd\VM Server-F_Drive.vhdx' to the destination file \\Hyper-V Host 2\Hyper-V Host 1.113901622$\f\temp\Virtual Hard Disks\VM Server-F_Drive.vhdx': 'Incorrect function.'('0x80070001').

Anyone seen this before?

↧

a disk read error occurred - after moving VM storage

July 28, 2013, 11:44 pm

≫ Next: Live migration VM with synthetic FC HBA fails

≪ Previous: Move-VM - 'Incorrect function.'('0x80070001').

Hi,

I have a VM that failed after moving VM storage in Hyper-V, and on reboot the server is now showing "A disk read error occurred. Press ctrl alt delete to restart". Today I moved several VMs, all others moved without issue.

I have tried a couple of recovery steps:

1 - CHKDSK - I get an error about corrupt master file table and then chkdsk exits.

2 - bootrec /fixmbr and bootrec /fixboot - no change after reboot

3 - attaching all the disks (1x IDE & 2x SCSI) to another VM - once brought online they show as RAW and containing no data.

The file sizes are still correct. What else can I do to attempt to recover data from this disk?

↧

Live migration VM with synthetic FC HBA fails

July 30, 2013, 1:57 am

≫ Next: Hyper-V servers

≪ Previous: a disk read error occurred - after moving VM storage

Hi all.

Yesterday I try to migrate my VM between 2 nodes and receive a error:

Live migration of 'Virtual Machine 'VMName' failed.

Virtual machine migration operation for 'VMName' failed at migration destination 'SecondServer'. (Virtual machine ID 22CF27B6-60F7-43C2-8DA2-7E16E9990642)

'VMName' Synthetic FibreChannel Port: Failed to start reserving resources with Error 'Cannot create a file when that file already exists.' (0x800700B7). (Virtual machine ID 22CF27B6-60F7-43C2-8DA2-7E16E9990642)

'VMName': NPIV virtual port operation on virtual port (C003FF2316FD0025) failed with an error: The world wide port name already exists on the fabric. (Virtual machine ID 22CF27B6-60F7-43C2-8DA2-7E16E9990642)

'50C885CE-AA51-4703-9F47-CC6710328795': NPIV virtual port operation for virtual port (C003FF2316FD0026) on physical port (50014380072C2B98) failed with an error: (0x800700B7) and NPIV error: (0x00000000). (Virtual machine ID 22CF27B6-60F7-43C2-8DA2-7E16E9990642)

VM work in 2 nodes cluster on Windows Server 2012 Standard. Each Cluster node has 2 port FC HBA. Each port connect to different physic FC Fabric. Each cluster node has 2 Virtual Fiber Channel SAN.

VM has next FC configuration:

SanName               : Fabric 1
WorldWideNodeNameSetA : C003FF2316FD0024
WorldWidePortNameSetA : C003FF2316FD0025
WorldWideNodeNameSetB : C003FF2316FD0024
WorldWidePortNameSetB : C003FF2316FD0026
ComputerName          : FirstServer
Name                  : Fibre Channel Adapter
Id                    : Microsoft:22CF27B6-60F7-43C2-8DA2-7E16E9990642\50C885CE-AA51-4703-9F47-CC6710328795
IsDeleted             : False
VMId                  : 22cf27b6-60f7-43c2-8da2-7e16e9990642
VMName                : VMName
VMSnapshotId          : 00000000-0000-0000-0000-000000000000
VMSnapshotName        :

SanName               : Fabric 2
WorldWideNodeNameSetA : C003FF2316FD0024
WorldWidePortNameSetA : C003FF2316FD0027
WorldWideNodeNameSetB : C003FF2316FD0024
WorldWidePortNameSetB : C003FF2316FD0028
ComputerName          : Firstserver
Name                  : Fibre Channel Adapter
Id                    : Microsoft:22CF27B6-60F7-43C2-8DA2-7E16E9990642\86F19D9D-464E-44C6-A117-4B7373A2AC69
IsDeleted             : False
VMId                  : 22cf27b6-60f7-43c2-8da2-7e16e9990642
VMName                : VMName
VMSnapshotId          : 00000000-0000-0000-0000-000000000000
VMSnapshotName        :

I see next WWPN from this VM machine: C003FF2316FD0025; C003FF2316FD0027 in SAN at now

Problems occurs only when I migrate VM from FirstServer to SecondServer. Migration VM (with FC VHBA) from SecondServer to FirstServer works fine.

Help me to understand this problem and step to resolve it.

Best Regards

↧

Hyper-V servers

July 22, 2013, 6:36 am

≫ Next: Recover a crashed Windows Server 2012 Cluster Storage Resource

≪ Previous: Live migration VM with synthetic FC HBA fails

Hi,

Is there a way to list all hyper-v servers (physical machines which are running hyper-v) from powershell? There is a way when all servers have at least one VM but what to do with does that do not have any?

Thanks.

↧

Recover a crashed Windows Server 2012 Cluster Storage Resource

July 30, 2013, 1:56 am

≫ Next: Error when creating VHDX via Hyper-V Manager remotely

≪ Previous: Hyper-V servers

Hi Guys,

Would much appreciate your urgent help on this. I've got a crashed Windows Server 2012 Hyper-V 4 node lab cluster environment. The cluster was deployed on a single DC based domain which unfortunately crashed last week. Because of the absence of a Active Directory, access to the Failover Cluster Manager MMC is not possible as I need to be logged in to the domain to have access to the cluster components. Now I need to recover the VMs on one of the nodes. The plan is to recover the CSV which is being hosted by a NEXSAN ISCSI SAN and then retrieve the VHDX files. I can still can access the LUN hosting the Cluster storage. However when I try to bring the disk online via disk management I get the following message"specified disk or volume is managed by the Microsoft failover clustering component. The disk must be in cluster maintenance mode and the cluster resource status must be online to perform this operation."

Any information on how to bring the cluster storage disk online for a broken cluster instance would be appreciated.

Regards,

Buchi

↧

Error when creating VHDX via Hyper-V Manager remotely

October 19, 2012, 5:45 pm

≫ Next: Virtual Machine Connection - Screen Resolution Does Not Match Guest OS - Cannot Resize the Window

≪ Previous: Recover a crashed Windows Server 2012 Cluster Storage Resource

Hey all!

So, I installed Hyper-V 2012 Core on a box, patched it, and then joined it to the domain. If I connect to the server via Hyper-V Manager from either a Windows 8 machine or Server 2012 machine and try to create a VM, it says "There was an unexpected error in configuring the hard disk on [Insert VM Name Here]." I feel like I've tried everything. I can not create the disk manually via the Manager on its own, before or after the VM being created. I can, however, create the disk via PowerShell with out a problem! Via PowerShell, I can create a dynamic VHDX disk every time without an issue.

I was curious if anyone has had this problem and/or knows a solution.

Thanks in advance!

↧

Virtual Machine Connection - Screen Resolution Does Not Match Guest OS - Cannot Resize the Window

September 26, 2012, 7:47 am

≫ Next: All VM's locked up in Server 2012 Cluster!

≪ Previous: Error when creating VHDX via Hyper-V Manager remotely

I am running several WinXP Guest VM's in Hyper V a 2008 R2Server.

All of the VM's run at display resolution 1024x768.

I have recently installed Win7 on my desktop management PC.

The problem I am experiencing is that, on certain VM's (not all), when I connect from my desktop PC to the Guest VM via the "Virtual Machine Connection" in Hyper-V Manager, the Virtual Machine Connection window does not show the whole desktop of the Guest VM.

Also Virtual Machine Connection will not allow me to resize the window to view the whole desktop of the Guest VM.

If I try and maxamize the Virtual Machine Connection window, it only switches my client PC to fullscreen, but still only displays part of the Guest VM screen.

See screenshot below of the issue.

Odly enough, when I click "Capture screen" inside Virtual Machine Connection, it captures the whole window and copies it see below.

I have search all over for this specific issue but I cant find any results.

Has anyone else experienced this?

Jackson Lancaster

↧

All VM's locked up in Server 2012 Cluster!

April 14, 2013, 2:02 pm

≫ Next: Hyper-V bin file consuming disk space.

≪ Previous: Virtual Machine Connection - Screen Resolution Does Not Match Guest OS - Cannot Resize the Window

Hi,

About 2 hours ago this evening, we lost communication to all 20 VM's on a 8 Node Cluster and they took 25 minutes to recover! Fibre Channel shared storage is being used throughout the cluster through 2 redundant routes from each node. The host OS was still working fine. I appreciate this points to the storage going down but I really don't think it had, there are no errors on either FC switch and no error on either storage device. We are using Server 2012 on HP DL360's.

The Get-ClusterLog returns this error at the time they all went down:

00000370.000010d8::2013/04/14-20:12:25.115 ERR [RHS] RhsCall::DeadlockMonitor: Call TERMINATERESOURCE timed out by 6 milliseconds for resource 'SCVMM VM1'.
00000370.000010d8::2013/04/14-20:12:25.115 INFO [RHS] Enabling RHS termination watchdog with timeout 1200000 and recovery action 3.
00000370.000010d8::2013/04/14-20:12:25.115 ERR [RHS] Resource SCVMM VM1 handling deadlock. Cleaning current operation and terminating RHS process.
000006dc.000011b4::2013/04/14-20:12:25.115 WARN [RCM] HandleMonitorReply: FAILURENOTIFICATION for 'SCVMM VM1', gen(3) result 4/0.
000006dc.000011b4::2013/04/14-20:12:25.115 INFO [RCM] rcm::RcmResource::HandleMonitorReply: Resource 'SCVMM VM1' consecutive failure count 1.
00000370.000010d8::2013/04/14-20:12:25.115 ERR [RHS] About to send WER report.

Any help would be appreciated.

↧

Hyper-V bin file consuming disk space.

May 26, 2009, 6:52 am

≫ Next: HIPER-V dynamic memory - assigning more maximum memory that phisycal

≪ Previous: All VM's locked up in Server 2012 Cluster!

Hi,

Host OS Configuration:- Windows 2008 Enterprise ,Hyper-V with KB950050 (64bit) update.
Host Hardware Configuration:- Xeon X7350@ 2.9 GHz x 4 CPU / 64 GB

I installed virtual servers on Hyper-V and found that everytime ".bin" file( with GUID) in snapshot folder ( default path is - C:\ProgramData\Microsoft\Windows\Hyper-V\Virtual Machines - this location can be changed using the snapshot location settings in Hyper-V settings) consuming disk space exactly same as memory configured for the guest OS in host system disk.It get disappeared when the guest OS became down.
we require to configure virtual servers with 20 GB RAM or more & found it requires additional disk same as RAM allocated in host system apart from the VHD allocating.

Does this mandatory ? i didnt find any MS article stating anything about it.
there is no snapshot created in hyper-v for virtual servers anyway.
Can we avoid this additional disk space requirement ?

↧

HIPER-V dynamic memory - assigning more maximum memory that phisycal

July 30, 2013, 4:18 am

≫ Next: Hyper-V Server 2012 Host RAM >2GB usage?

≪ Previous: Hyper-V bin file consuming disk space.

The physical machine has 16GB of RAM

Can I configure two VM’s with dynamic memory allocation like this?

4GB start; 12GB Maximum; priority low
2GB start; 14GB Maximum; priority high

What about the memory required by the host (~2GB)? Is there a way to "fix" the host memory to ensure that it has the necessary ammount to run properly?

I know this is not necessary a best practice, but I have one machine that - every two hours – performs a highly memory demanding process and then becomes almost idle and the second one, that works constantly, will benefit from the extra memory in the time between.

↧

Hyper-V Server 2012 Host RAM >2GB usage?

July 29, 2013, 9:12 pm

≫ Next: Disk space required for Hyper-V 2008 R2 snapshot delete and merge

≪ Previous: HIPER-V dynamic memory - assigning more maximum memory that phisycal

I have a 4GB Hyper-V Server 2012 Host in a lab that uses a full 2171 MB for the Host. I have verified the usage via setting dynamic RAM up on the VMs to see how much they could get. I know the Host is woefully low on RAM for a Hyper-V Host but I am still curious, is this number right? Does Hyper-V Server 2012 (free version, i.e. even lighter than Core install) require >2GB for the Host? All the docs seem to suggest roughly 1GB for Host plus a fee 10's of MB per VM but nothing says it should be >2GB for a handful of VM's in total. I want to know for sizing other Production deployments as well as to know if there is tuning to be done that we are missing.

↧

Disk space required for Hyper-V 2008 R2 snapshot delete and merge

July 30, 2013, 8:20 am

≫ Next: Hyper-v hardware utilization/performance

≪ Previous: Hyper-V Server 2012 Host RAM >2GB usage?

In Server 2008 R2 how much free disk space is required to delete a Hyper-V snapshot and merge it into the VHD file? Do I need free space equal to current VHD + the snapshot files or just free space equal to the snapshot files? For example if my VHD is 100 GB and the snapshot folder is 25 GB do I need 125 GB of free space or would 25 GB of space be sufficient? If the VHD is completely re-written it makes sense that I would need 125 GB of free space. If the VHD file is simply appended to I would assume that 25 GB is all that would be needed.

I know that this has been asked before but I wanted to get a little bit more clarification. Thanks for your help.

↧

Hyper-v hardware utilization/performance

July 30, 2013, 8:29 am

≫ Next: Live Migration half as fast depending on source node

≪ Previous: Disk space required for Hyper-V 2008 R2 snapshot delete and merge

I need a little clarification on how Hyper-v utilizes the hardware, particularly the CPU.

If I understand this correctly -and correct me if im wrong- the hypervisor uses a round-robin approach to give all VM’s equal CPU time. So for example, say I have 2 quad-core processors (socket A and B) with HT support, meaning, 8 cores and 16 logical processors. I setup 2 VM’s and assign each VM 16 VP’s (overcommitting), the hypervisor will first give VM1 access to all 16 LP’s, then take it away from VM1 and give all 16 LP’s to VM2, then back to VM1 and the cycles continues in an equal round-robin fashion (unless I increase or decrease the relative weight of a given VM).

So I’ve got a couple of questions:

1. What if I assign only 8 VP’s to each VM, when VM1 is given CPU time, it only uses half the CPU load (8 LP’s). Does this mean that both VM1 and VM2 will/can now continuously be given equal CPU time at the “same” time?

2. If the above is not true, when I turn on VM1 with its 8 VP’s, it (say) grabs LP 16-8. Let’s say VM1 is a very high CPU demander, keeping (all 8 LP of) the CPU spiked to 90%. The remaining LP’s (8-1) are practically idle. When VM2 is given CPU time (for its 8 VP’s), which LP’s will it use? How does the hypervisor ensure that VM1 can continuously meet its 90% usage demand without affecting VM2?

3. What is considered healthy for the Context Switches (Hyper-v Hypervisor Logical Processor) counter?

4. Is there a “proc queue length” counter for the VP? If not, is the System\Processor Queue Length counter reflecting the VM or only the parent partition? If not, how can I monitor or determine whether a VM has a bottleneck/queuing issue?

5. And one last question about the network. Is the \Network Interface\Output Queue Length and Network Interface\Bytes Total/sec counters (when adding the counter only for the physical NIC that was added the virtual switch) reflecting the reality of what the VM’s are experiencing?

↧

Live Migration half as fast depending on source node

July 29, 2013, 4:06 pm

≫ Next: v2v from vSphere 5.1 to HyperV

≪ Previous: Hyper-v hardware utilization/performance

So this is pretty weird. Brand new pair of Dell R620 servers with Broadcom 57800 10GbE NICs set up in a cluster. All Windows Updates and NIC driver updates installed. If I Live Migrate from Node 1 to Node 2, I see >9.5Gb/s transfer speeds, which means that the 7 VMs I have running migrate over in about 17 seconds. Not too shabby. If I migrate the VMs back, I only see about 5Gb/s speeds, and it takes about twice as long: 34 seconds. Both servers' Hyper-V Manager are set for up to 8 simultaneous migrations. I can't find any difference in settings between the two. I'm stumped! Any ideas?

↧

v2v from vSphere 5.1 to HyperV

July 30, 2013, 10:58 am

≫ Next: VMM 2012 sp1 system job - refresh virtual machine

≪ Previous: Live Migration half as fast depending on source node

I have just embarked on a project to v2v 250 VMs from vSphere 5.1 to Hyper-V 3.0. I thought I'd start a discussion string on v2v experiences. I have used the native v2v conversion with in VMM, 5nine's Easy Converter, Starwind's Converter and MVMC. I'm not happy with any of them. They are so inconsistent and frustrating, I see this project as literally taking years. Sometimes the VMM converter works, sometimes it just fails with crypt error messages. At the beginning it would crash the management interfaces on the ESXi servers but a ticket with VMware and increasing some system values solved that.

Most of the time, it will run for an hour or two, then just fail with no explanation. Same for MVMC. Occasionally I have some luck doing a p2v instead of a v2.

5nine usually fails, but gives me the VHDX files so I can create the VM from there. Ditto for Starwind.

Even if I could get a product that works consistently, the fact that 5nine and VMM actually make 2 or 3 copies of the VMDKs during the conversion process is pretty ridiculous. I have had to carve out terabyte LUNs to attach to a Hyper-V host just to have enough scratch disk to convert, then watch the conversion take literally hours.

All in all, a very frustrating experience. Have others been faced with similar issues? And if so, how did you tackle it?

Thanks.

↧

VMM 2012 sp1 system job - refresh virtual machine

July 30, 2013, 1:41 pm

≫ Next: Get-childitem with recurse and create folders in a different directory

≪ Previous: v2v from vSphere 5.1 to HyperV

Last night around 1030pm the service account for vmm 2012 kicked off a "Refresh VM - System Job" that refreshed the vm NIC configuration. It just so happens that a different NIC was assigned to the vm breaking the application license on the vm. Does anyone have any links/docs/info/insight into how to manage a System Job or stop them from occurring?

↧

Get-childitem with recurse and create folders in a different directory

July 24, 2013, 2:09 am

≫ Next: Failover Cluster WMI Provider detected an invalid character - W2008R2 SP1

≪ Previous: VMM 2012 sp1 system job - refresh virtual machine

I've got a folder containing child folders which I am doing a get-childitem -recurse on. My script at the moment is zipping the top level file in the script but I need it to do a little more.

For starters I need to zip all the files in each folder found but to a mirrored directory\archive somewhere else on the file system.

I have not got any code yet, does anyone know how I would make a start on this?

Alter De Ruine

↧

Failover Cluster WMI Provider detected an invalid character - W2008R2 SP1

July 29, 2013, 6:44 am

≫ Next: Hyper-V Network Adapter Error 31 After Migration

≪ Previous: Get-childitem with recurse and create folders in a different directory

We have a four node Hyper-V cluster running Windows 2008 R2 SP1 Data centre edition

For several months the following error has been flooding the application logs on the servers.

Failover Cluster WMI Provider detected an invalid character. The private property name 'PreviousOfflineAction' had an invalid character but the provider failed to change it to a valid property name. Property names must start with A-Z or a-z, and valid characters for WMI property names are A-Z, a-z, 0-9, and '_'.

The exact issue is detailed here with a Hotfix http://support.microsoft.com/kb/974930

Unfortunately the Hotfix is not applicable / won’t install on Windows 2008 R2 SP1

I've seen a few articles mentioning SCVMM, we run SCVMM 2012 SP1 it seems none of the fixes I've found so far relate to that.

Any ideas on how I can eliminate this error?

↧

Hyper-V Network Adapter Error 31 After Migration

July 24, 2013, 12:14 pm

≫ Next: Hyper-V Server Manager -Error occurred during enumeration of Data Deduplication volumes

≪ Previous: Failover Cluster WMI Provider detected an invalid character - W2008R2 SP1

Migrating servers from 2008r2 Hyper-v cluster to new 2012 Hyper-v Cluster. All 140 servers have migrated with no issues until my last one.

It is our secondary domain controller. It is running 2008r2 fully patched. I have copied the vhd to the new cluster storage and created the vm profile in 2012 failover cluster manager.

Server boots and I have installed newest integration drivers. The network connection goes to Red X. In device manager I have an exclamation mark on the network adapter. Error 31 can't load drivers.

If I shut it down and boot it back up on our 2008r2 Hyper-v cluster, networking works as expected.

Are there some standalone drivers that I can manually install on the vm or another workaround?

I have tried registry hacks in HKLM\SYSTEM\CurrentControlSet\Control\Network delete theConfig (REG_Binary) key, also tried removing adapter and reinserting, removing adapter and adding legacy adapter. Same results. I am out of ideas and google search reveals no similar issues so far.

Anyone experienced this?

↧

Latest Images