Requirements for backup Hyper-V 2012 VMS

April 15, 2014, 10:41 am

≪ Previous: SBS 2011 offline when VSS Backup runs

Hello Everyone,

I have implemented Symantec Netbackup 7.6.0.1 to backup Hyper-V 2012 VMS. We have Hyper-V 2012 (RTM) 3.0 .

In some cases the virtual machines enter in saved state. Following Netbackup documentation for Hyper-V and after do some research, consolidate all information in a spreadsheet, see below...

Guest OS compatibility;
Integration Service is installed and updated;
Guest OS updated;
VM has only basic disk;
Shadow copy must be available to Hyper-V VSS and be located at the same volume;
VMs have at least 20% of free space on each disk;
VSS infrastructure is health and dos not present any issue;
WMI is OK;
At time of analysis, any peak of processing was presented;

I have worked with Netbackup, Backup Exec and Simpana and in all documentations, the type of backup (ONLINE vs OFFLINE) is a Microsoft designation and the backup application can't decide about how to backup a virtual machine.

After check all pre-requirements present above 99% of issues related with vms enter in saved state when backup happens was solved.

But we still have some machines (4) enter in saved state, VMS is 2008 R2 with SP1 all VMS are in compliance with pre requirements table, BUT all VMS isn't compliance with item number 9. For testing this VMS I do a clone, because they are not in production the resources is dropping and the clones no more enter in saved state, so I use some .vbs scripts for stressing the VMS, and after stress, the VM enter in saved state.

Have Microsoft some best practices for applying in Virtual Machine, like "20% of free resources when backup operation is run"...

Thank You!

↧

Hyper V guest slow disk access

August 6, 2013, 1:53 pm

≫ Next: About changing the subnet mask. Help me please

≪ Previous: Requirements for backup Hyper-V 2012 VMS

I have a Hyper V cluster(2012) with a 2012 guest running 2012 acting as my file server.

My 2 hosts are connected to a gig switch stack which in turn is connected to my Dell MD3200i.

My file server has a "data" disk that is a VHD(fixed) on the iSCSI storage. As is it OS(dynamic)

I'm coping 200GB right now and only seeing data thoughtput of 8MB/sec.

When I check perfrmance on the file server the disk queue is huge. But when I check the queue for C: on the host, which is where the cluster storage appears to sit, its got a low if not no queue.

Also did a copy using the host and saw 58MB/sec so I'm happy the network should be able to pass the data.

Switch CPU is at 5%

Can anyone tell me what I'm doing wrong?

↧

About changing the subnet mask. Help me please

April 17, 2014, 6:26 am

≫ Next: Server 2012 core

≪ Previous: Hyper V guest slow disk access

Hi guys

We have to increase the number of XenApp servers drastically which leads the IP address in our PVS LAN being not enough. Our method is to change the subnet mask of PVS LAN dhcp scope so that we can get more IP addresses to use. However I have to evaluate the risk of doing it. I need your help.

I'd like to introduce our IT environment to you first. We have a cluster with tens of Hyper-V hosts. We have 2 PVS servers, about 50 XenApp servers(for public desktop) and other management servers. There will be about 6,000 users using public desktops delivered by XenApp servers. And all the XenApp servers are connected will Business LAN and PVS LAN (the target of this work). Now I have to change the subnet mask of all the XenApp servers(change only happens on PVS Nic), the PVS servers(change only happens on PVS Nic) and also the hosts.

My plan is Step 1: I will change the mask subnet of all hosts first by using vmotion function. I mean I will migrant the VMs from host A to host B and change the subnet mask of host A, then get ex-host A VMs back meanwhile move the VMs on hosts B to host A so that I can do the same thing on host B. Here is a problem. If I changed the subnet of host A(just NIC for PVS), can I really get ex-host A VMs back from host B?

Step 2: I am gonna change the subnet mask of my Slave PVS server. I think it will not affect anything.

Step 3: As I did in step 1, I plan to change subnet mask of XenAPP servers by similar method. I mean I will cut the network of XenApp server 1 off so that users will use other XenApp servers except XenApp server 1. Then I can change the subnet mask of XenApp server 1 with no user using this server. The rest can be done in same manner I think.

Step 4: Finally I will deal with the main PVS server.

Certainly I have to change the subnet mask of the rest management servers by turns.

I am not sure if you could understand my plan. If you could or have such experience, put judge it for or give me some advice.

Thank you

↧

Server 2012 core

April 16, 2014, 9:35 am

≫ Next: Mouse not captured in remote desktop session

≪ Previous: About changing the subnet mask. Help me please

Hi all,

We are using the above for hyper V. I believe it might be the Hyper V free installation. Im trying to add the domain admins group to the local administrator by going to option 3 then specifying "Domain\Domain admins" I get the error "there is no such global user or group: domain\DOMAIN. there is no such global user or group: domain\ADMINS.

So i thought it was the space causing the issue so i have tried it with space and with domain^admins but no luck. Can anyone point me in the right direction. Im guessing its something real simple that im missing. I have added just myself into local admin (option3) and it worked fine. The server is on the domain. Thank.

↧

Mouse not captured in remote desktop session

April 17, 2014, 7:17 am

≫ Next: Hyper-V 2012 R2 & Windows 8.1 VMs performance metrics using WMI query

≪ Previous: Server 2012 core

I have one (but only one) Hyper-V VM which has the "mouse not captured in remote desktop session" issue. I have researched this problem extensively and so far none of the suggested fixes I have found are working.

The VM is running Windows Server 2012. It is completely up-to-date.

The Hyper-V server its running on is running Windows Server 2012 R2. Also up-to-date.

All integration services are offered in the VM according to the Hyper-V management tool. I have tried removing all integration services from the VM and putting them back with the Hyper-V management tool. This had no effect.

I have also set the VM to "Detect HAL" using BCDEdit... this had no effect.

The only suggestion I could not try is the suggestion to use the "Insert Integration Services Setup Disk..." command from the "Action" menu. I can not do this because a message to "Tap to choose what happens with this device" message pops up when I try it... and, guess what? I can't "tap" it because like everyone else we don't use touch screens in the server room. I can't click it either because my mouse doesn't work. Trying to click it just brings up the same "mouse not captured in remote desktop session" message I am trying to get rid of.

Any suggestions?

↧

Hyper-V 2012 R2 & Windows 8.1 VMs performance metrics using WMI query

April 17, 2014, 7:22 am

≫ Next: Hyper-V VMs memory utilization

≪ Previous: Mouse not captured in remote desktop session

Hi,

I used the following query to fetch the Disk utilization for Hyper-V VMs (non v2 hosts) using its host credential "select ReadBytesPersec,WriteBytesPersec,ReadCount,WriteCount from Win32_PerfRawData_StorageStats_HyperVVirtualStorageDevice" but for the V2 (Windows 8.1 & Windows 2012 R2) this query is not working this is throwing Invalid class using the namespace "root\cimv2".

Please let me know the exact WMI class for the Disk utilization for V2 hosts VMs.

Regards

Raamesh Keerthi N J

↧

Hyper-V VMs memory utilization

April 17, 2014, 7:24 am

≫ Next: Question on backup network link

≪ Previous: Hyper-V 2012 R2 & Windows 8.1 VMs performance metrics using WMI query

Hi,

Please let me know the WMI class (query) to check the memory utilization for all the VMs under the Hyper-V host.

Thanks & Regards,
Raamesh Keerthi N J

↧

Question on backup network link

April 17, 2014, 7:30 am

≫ Next: DSRM black screen issue with 2012R2 DC running on 2012R2 Hyper-V

≪ Previous: Hyper-V VMs memory utilization

I created an internal network using Hyper-V on a 10.x.x. network. I want to cluster 2 of the servers, so I created an additional NIC on the 2 servers. I put the second NIC on the same subnet. I can ping Server 1 from Server 2, but can't ping Server 2 from Server 1.

I tried put the second NICs on a different subnet. 192.168.1.1 and 192.168.1.2. I got the same results.

I know this is not a best practice, but can anyone shed some light on this? BTW, The NICs are using the same internal network .

Any help appreciated.

scott

Scott D Hastings, Sr.

↧

DSRM black screen issue with 2012R2 DC running on 2012R2 Hyper-V

April 16, 2014, 2:57 am

≫ Next: Error with Hyper V snapshot

≪ Previous: Question on backup network link

Hi!

I have been testing AD disaster recovery with few domain controllers (Server 2012R2), running on Server 2012R2 and Hyper-V. Everything OK unless I try to log in to Domain Services Recovery Mode.
After username&passwd screen goes all black and nothing happens. Every time, every server (tested this with 3 different HP workstations, one HP ML350G6 server and one laptop). Hyper-V virtual machines tested both gen1 and gen2.
With 2008R2 Hyper-V and same domain controller on it everything works fine. This makes systemstate restore bit more difficult.

Am I missing something or is it just a bug? Googling didn't help this time.

↧

Error with Hyper V snapshot

April 17, 2014, 7:57 am

≫ Next: hyper-v 2012 missing vhdsvc service

≪ Previous: DSRM black screen issue with 2012R2 DC running on 2012R2 Hyper-V

when I run vssadmin list writers this command I get this error:

Writer name: 'Microsoft Hyper-V VSS Writer'
   Writer Id: {66841cd4-6ded-4f4b-8f17-fd23f8ddc3de}
   Writer Instance Id: {5fd999d8-3a4b-4fbd-81ba-27bc0ab5c203}
   State: [8] Failed
   Last error: Non-retryable error

I can not create snapshot with Hyper V, VSS restart services and is still the error when i run the vssadmin list writers

↧

hyper-v 2012 missing vhdsvc service

April 17, 2014, 8:18 am

≫ Next: Move VM from server with DC to server with no DC

≪ Previous: Error with Hyper V snapshot

I have a new Hyper-V server that is missing two services

vhdsvc and nvspwmi

How can I get these two pieces installed?

↧

Move VM from server with DC to server with no DC

April 16, 2014, 3:56 am

≫ Next: [Forum FAQ] Resolve virtual machines fail to start issues

≪ Previous: hyper-v 2012 missing vhdsvc service

I need to move my newly created VM's to new server that does not have a DC. I run two web servers, small business and do not need a DC environment. My Web servers are just the way I want them , I did not want to rebuild everything so I did a P2V via the SCVMM SP1 instructions.

I set up a DC on a VM to ran the SCVMM SP1 and now have my VM's that work great on the Host when logged into the Domain.

PROBLEM:

When I log the host off the Domain the VM's are no longer available.

If I log the host computer back onto the Domain, the VM's are both available.

1: I can not import the VM's into the Hyper V manager unless Host is logged on Domain - WHY.

2: The VM's should run on the host without having to be logged onto the Domain.

3: How do I disconnect the VM's from the domain to run on a workgroup.

One VM's is logged on to the domain, that is the DC VM. The other is not and never was.

Problem is the Hyper V Manager requires I be logged onto domain to see VM's.

Anyone have a simple answer.

Thanks

↧

[Forum FAQ] Resolve virtual machines fail to start issues

April 17, 2014, 9:23 am

≫ Next: All Windows VM - Virtual Disk Service start/stop every minute

≪ Previous: Move VM from server with DC to server with no DC

After you restart the Hyper-V Host server and the guest virtual machines, you may encounter the error:

An error occurred while attempting to start the selected virtual machines(s).

Here are three common issues.

Issue 1

On a Windows Server 2008-based machine that has the Hyper-V role installed,or on a Microsoft Hyper-V Server 2008-based machine, you encounter the following:

• When you try to connect to a virtual machine, error occurs:

Cannot connect to the virtual machine because the authentication certificate is expired or invalid. Would you like to try connecting again,

• When you start a virtual machine, error occurs (Figure 1):

An error occurred while attempting to change the state of virtual machine ‘VMname’

'VMName' failed to initialize.

Could not initialize machine remoting system. Error: ‘Unspecified error’ (0x80004005).

Could not find a usable certificate. Error:‘Unspecified error’(0x80004005).

Figure 1: ‘Unspecified error’ (0x80004005).

Cause:

This issue may occur if the Hyper-V Virtual Machine Management service (VMMS) certificate has expired.

Solution:

Install the Hotfix.

You cannot connect to a virtual machine when the Windows Server 2008 Hyper-V VMMS certificate has expired

http://support.microsoft.com/kb/kb967902/en-us

Workaround:

Follow the steps below to generate a new certificate.

1. Shut down or save all virtual machines.

2. Restart the Hyper-V VMMS.

3. Start the virtual machines.

Note:As the Hyper-V VMMS certificate is valid for one year, the update has to be applied to prevent this issue from occurring in the future.

↧

All Windows VM - Virtual Disk Service start/stop every minute

April 14, 2014, 11:42 pm

≫ Next: Hyper V server 2012 R2 network interface index number mismatch

≪ Previous: [Forum FAQ] Resolve virtual machines fail to start issues

Dear all

I recently discovered that all my Windows VM's (2003/2008/2008R2) start and stop the virtual disk service:

System Log File:

08:37:27 Virtual Disk Service --> Service started
08:37:27 Service Control Manager --> The Virtual Disk service entered the running state.
08:37:28 Virtual Disk Service --> Service stopped
08:37:28 Service Control Manager --> The Virtual Disk service entered the stopped state.

and i have this entry every minute.

I removed the DVD Drive from the VM but no change. What could be the reason for this? All my physical servers with the same backup program haven't this behavior.

Thanks for any hint

JBAB

↧

Hyper V server 2012 R2 network interface index number mismatch

April 17, 2014, 9:56 am

≫ Next: Best way to set up a virtual lab

≪ Previous: All Windows VM - Virtual Disk Service start/stop every minute

I have a hyper v server 2012 R2 that has a mismatch on the network adapter index number between the sconfig console and the actual index number. If i go into the network settings in sconfig i see an index number of 17 for the Hyper-V virtual Ethernet Adapter #2. But if i select that index sconfig crashes out to a command prompt. If i use netsh to query the interface the index number is 18. If i try to use the 18 index number in sconfig then it also crashes out to a command prompt.

I am not sure when this discrepancy occurred. The server has been running for well over a year and no network changes have been made since initial setup. I suspect that it happened during an update.

Is there any way to fix this? would be nice to be able to use sconfig to update static dns servers. doing it via netsh is a pain.

↧

Best way to set up a virtual lab

April 17, 2014, 11:40 am

≫ Next: Hyper-V with Windows Server Essentials in a VM - adding a second physical disk to VM - now VM will not start

≪ Previous: Hyper V server 2012 R2 network interface index number mismatch

Hi guys

I now this have been answered a lot of times before, but somehow i can get it to work

What i want is to build a internal network in my Hyper-V with also have access to the internet

This is how i i have configured

My host have just one nic

But i have created two swiches(one internal and one external)

When i started by setting up the DC(For the internal network)

Configures to use the virtual switch

IP: 10.10.10.2

Mask: 255.0.0.0

Gateway: 10.10.10.1

Gateway

Configured two nic, one to use the internal, and the second to use the external

The nic configured with the external switch is set to DHCP. This is used if i trave with my machine i always can connect to the internet with my lab environment

The second nic is configured to use internal switch

IP: 10.10.10.1

Mask 255.0.0.0

Gateway: None

No i try to install RRAS on the server i guess i need to use NAT here

When i now switch to my DC i am able to ping the gateway i am able to ping the other gateway also 192.168.1.1.

But i am not able to connect the internet. From my gateway everything is working just fine

Hope someone can help me answer what im missing

↧

Hyper-V with Windows Server Essentials in a VM - adding a second physical disk to VM - now VM will not start

March 23, 2014, 8:03 pm

≫ Next: avhdx file growing

≪ Previous: Best way to set up a virtual lab

Hardware previously had Windows Server Essentials 2012 running fine.

Current setup, base is Hyper-V 2012 with Windows Server Essentials in a VM, which has been installed a few weeks ago and is stable, but only on the main SSD.

I decided to try to configure so that the physical disks were available directly to WSE, hoping that the management of them would be easier.

I had one SCSI in the VM, and aside from the SSD boot disk, I was able to add 1 3TB drive, and it all works. If I add a second 3TB drive, the WSE VM will not start. It tries to start, but goes back to off in a few seconds.

I then decided to put the 3TB drives on a second virtual SCSI within the VM, to see if that was better.

No change, one drive ok, but two and the VM will not start.

Then I booted the VM with the 1 3TB drive that worked, and then while the VM was on, I was able to add 3 more 3TB drives, and the VM stayed up. I was able to create a storage space from the 4 drives, mirrored, thin provisioned, the standard WSE 2012 settings (I kind of wish it would give you a choice on these, but...). Everything seemed to be working fine.

But, when the VM was stopped, it will not now restart. I suppose I have to delete the 3 extra drives, which will cause storage pool disaster I presume.

Seems odd that I can add them, and use them in WSE2012, but then it will not restart. And they were on a separate virual SCSI controller from the "boot" drive.

Any ideas? (or let me know if this is the wrong forum... since the VM would not start, I am blaming Hyper-V...)

Thanks!

Ross

↧

avhdx file growing

April 17, 2014, 11:00 am

≫ Next: Windows 2012 R2 Hyper-V - Guest network very slow

≪ Previous: Hyper-V with Windows Server Essentials in a VM - adding a second physical disk to VM - now VM will not start

Hello

Server 2012 with hyper-v

2 Virtual servers (one DC/File server)

On the D drive, I have allocated 4.7 TB of diskspace

I have a few avhdx files. I was restoring files from a crash last week (because the Host ran out of space). I recovered data to the File server (after making a TB of space on the Host). After copying files into place on the file server, the avhdx file grew to over a TB. When I deleted files on the File server, the avhdx file did not shrink as I expected it to... so it threatens now to do the same thing as last week.. I currently only 46 GB of HD space on the Host.

I desparetely need to know how to fix this. I know for a fact that I have cleared at least .5TB off the File Server (guest).

Any suggestions?

Thank You

Terry

↧

Windows 2012 R2 Hyper-V - Guest network very slow

April 13, 2014, 5:46 pm

≫ Next: Hyper-V 2012 R2 and Windows 7 sp1 Enterprise Guest with vGPU

≪ Previous: avhdx file growing

Hi,

I'm having this very strange slow network problems on a Windows 2012 R2 Hyper-V cluster.

In the process of troubleshooting, I've simply the setup to the following:

2 Physical node, with one 4 ports Intel 1Gbps card and on 2 ports Intel 10Gbe card in each node. The nodes are setup in an Hyper-V cluster running Windows 2012 R2.
The 2 10Gbe ports are used for the cluster communications.

2 of the 1Gbps port are connected to a virtual switch (completely separate switches), so I have two Hyper-V virtual switch I can use to connect virtual machines (I removed teaming from the original configuration to make sure that wasn't the problem).

Here is the problem:
If I create a new virtual machine (windows 2012 R2 or Windows 2008 R2 guests - tested both) and connect it to the first virtual switch (let's call it Switch A), If i then copy a file from that virtual machine to another physical host on the network (outside the cluster), I get a normal 110MB/s transfer rate (no problem here). From the same guest, If I then start a file copy to another virtual machine (outside the cluster to either Hyper-V or VMWare VMs), I barely get 10MB/s!

Here is the bizarre thing: if I now connect the Guest to the other virtual switch (Called it Switch B), the transfer rate are inverted: I get 110MB/s if I copy to any VM, and I get 10MB/s if I copy to any physical server!

So using Switch A, I get great network performance to physical hosts, while crap performance to VMs, while using Switch B, I get the total opposite!

The two ports used are located on the same physical network card (4 ports card), with the same firmware and driver (latest from Intel)!

I'm lost... Anyone has seen this before?!

Thanks,
Stephane

↧

Hyper-V 2012 R2 and Windows 7 sp1 Enterprise Guest with vGPU

April 18, 2014, 4:07 am

≫ Next: Improper Install of Hyper-V 2012 R2 Integration Services on 2008 R2 VMs

≪ Previous: Windows 2012 R2 Hyper-V - Guest network very slow

We have just installed a new Hyper-V Server 2012 R2 with an NVIDIA GPU (K4000) that is recognised OK by Hyper-V.

When we move a Windows 7 sp1 Enterprise (with RDP 8.0 update) to the new server and assign the RemoteFX vGPU we can't connect to the VM via RDP. If we remove the vGPU we can then RDP to it OK.

I've had a look at the forums and there are several TechNet Blogs that suggest the Windows 7 Enterprise VM should work with the vGPU (http://blogs.msdn.com/b/rds/archive/2012/10/09/remote-desktop-protocol-8-0-update-for-windows-7-sp1-enabling-a-great-wan-user-experience-for-windows-7-sp1-virtual-desktops.aspx)

Anyone got any ideas what we're missing or doing wrong as we really need to get the VMs running with the vGPU ASAP.

Cheers for now

Russell

↧