Hello,

one of my customers asks whether to use differencing disks in very complex environment: 2000 clients, 24/7 environment, several dozen VM's. Personally I don't have any experience regarding this scenario. I'd rather be safe than sorry, anyway if anyone uses differencing disks in such complex environment please share your experience.

Lalas

Hi Community<o:p></o:p>

<o:p> </o:p>

We have multiple servers running Windows Server 2012 R2 Datacenter with Hyper-V Role installed, some Stand Alone and others in Failover Cluster.<o:p></o:p>

we are trying to live migrate a virtual machine and detect that the live migration takes forever, the troubleshooting report us that the traffic of the live migration is being send via the two network interfaces that we have in our servers.<o:p></o:p>

Our environment is the following, each server has 4 10Gb Physical NICs, we have 2 teams managed by SCVMM 2012R2, all interfaces are Trunk VLAN based.<o:p></o:p>

In the first team we create 2 team interfaces with different Vlans id (Management VLAN 100 10.0.0.0/24 and Migration VLAN 200 172.16.0.0/24) both interfaces has JumboPackets (MTU 9000) and we have perfect connectivity between the servers. <o:p></o:p>

The Hyper-v LiveMigration configuration is set to use the VLAN 200 (172.16.0.0/24) in each server. the active directory and DNS work only on the Management Network so the migration interfaces have unchecked the Register DNS checkbox, LMHosts and Netbios disabled.<o:p></o:p>

Here is the output of the NICs monitoring.<o:p></o:p>

and here is the Performance monitor Network Activity output<o:p></o:p>

this is a strange issue and we don’t have any idea of what could be happening here.<o:p></o:p>

<o:p> </o:p>

we'll appreciate all the help that you could provide us.

Excuse myEnglish, its not my native language.

Best Regards.

Julian Galeano

Cloud Architect Leader.



hello we have a bunch of VMs on standalone 2012 R2 Hyper-v with local storage.

i would like to migrate the VMs to a new 2 host cluster (2016) with iSCSI shared storage

can i add all the hv hosts to hv manager and just move them? what network will be used?

Hello,

I have :

1 Physical server with Server 2016 with the hyper-V role.

2 VM's, 1 RDSH (2016) and 1 RDSH(2016 with remoteFX

I set up a Remote Desktop deployment.

In the physical server is a Quadro GPU present. On 1 server I added a Remotefx 3d videocard with 1 GB memory.

Now the problem:

When the remotefx 3d videocard is present i can RDP to the server without a problem. But the remoteApps I publish just don't work, it start the mstsc.exe client and start connecting and just stops.

When I remove the remotefx video adapter from the VM and boot again then the remoteApps do work.

I searched through eventviewer for a cause but couldn't find anything. I searched through Microsoft docs if this setup is supported, the only thing I found was that a server with remotefx only supports 1 concurrent user. But found numerous forums that say that remoteApp should be supported combined with remotefx.

I'm thinking I just missing something crucial.

tl;dr: VM virtual disks stored on physical hosts disks are extreamly slow when tested from inside the VM. Test from the host they appear fine. 180 MiB/s vs 0.11 MiB/s

I have two identical physical Dell PE R640 servers, With internal storage. Internal Storage is PERC H740P

Both Servers have 3 Virtual disks: 
Disk 1, SSD-SATA Raid 10 , 8 x 1787GB SSD disks (mounted to the host as E:\)
Disk 2, HDD-SAS Raid-5, 4 x 1862GB SAS HDD (mounted to the host as G:\)
Disk 3, OS raid-1 SSD 111 GB (mounted to the host as c:\)

These Servers have been installed with windows server 2016 Standard and Hyper-V. 2 Virtual Machines on each physical server, The virtual machines are all also 2016 standard.

The virtual machines have 4 virtual disk drives, all VHDX's are fixed in size

C:\ stored on the HDD volume G:\ of the host
D:\ stored on the SSD volume E:\ of the host
E:\ stored on the SSD volume E:\ on the host
G:\ stored on the HDD volume G:\ of the host

All servers have been updated to the latest microsoft patches, the firmware, drivers, Bios, Storage controlers and Lifecycle controlers have been updated to the latest versions as reccomended by Dell product support. Im currently working with Dell product support on this issue

When we first Setup these servers everything worked as expected disk access from the VM's was normal. At some point before Christmas, both VM's on one of the hosts dropped their disk perfromance from super fast to super slow, the other two vm's on the other physical server remained super fast. 

We have been troubleshooting since then when all of a sudden the remaining two VM's dropped disk speed yesterday from supper fast to super slow. 

Whats the difference, here are the numbers from diskspd.exe, I used the same optoins in all tests
diskspd.exe -d15 -o4 -t4 -b8k -r -L -w50 -c1G testdiskspd.dat

https://pastebin.com/gKQvKTTG

Above is the SSD Storage of one of the two hosts benchmarked with diskspd.exe right now: some exerpts from that file:

Total IO = ~371 MiB/s
Read IO = ~180 MiB/s
Write IO = ~180 MiB/s

Total IO = ~47565.65 I/O per s
Read IO = ~23757.19 I/O per s
Write IO = ~23808.46 I/O per s

Now im going to perform the same test on one of the VM Guests, Neither of these VM's are currently performing any operations on the disk other than having been booted up. This test is performed from the VM, on its d:\ disk which is stored on the above disk

https://pastebin.com/QE7r2XNg

The same details as above

Total IO = ~0.21 MiB/s
Read IO = ~0.11 MiB/s
Write IO = ~0.10 MiB/s

Total IO = ~27.06 I/O per s
Read IO = ~13.93 I/O per s
Write IO = ~13.13 I/O per s

These results are consistent across both disks in the VM and on both Virtual machines hosted. Whats crazy is I did this test on one of the virtual Guests last week and its performace was almost identical to that of the host, and when this set up was first created, the performace of all the VMs was almost exactly that of the host. and now its almost at a standstil. 

Im at a loss of where to go from here, Its not something ive seen before so any help woudl be appriciated. thanks

Anybody knows the structure of this file? And wants to share?

Seb

Running my testing of conversion from shared vhdx (part of cluster) to vhd set (on Server 2019)

Each time the conversion finished fine, but end product is unusable set (it is not access from within 2012 R2 OS (VM)

If I do attach only renamed .avhdx , it shows perfectly fine (as NTFS)

Anybody having any idea what is wrong with the SET?

If I create brand new set, it works fine, it is only converted set that does not

Seb

Microsoft will not handle this as 2012 R2 is no longer supported.

I downloaded an evaluation of Hyper V server.

I thought it would install as a service on WIndows Server 2012 R2 Std. 

Instead it installed Hyper V Server replacing WIndows Server.

Couple questions ... 

Can the Hyper V Server installation be reversed.

Can I access a browser on the Hyper V server.

Right now all we see is a Task Manager.  If Task Manager is closed then we just have a black screen.

No apparent way to execute applications. 

Need help big time.

I have responsibility for multiple large-ish Hyper-V failover clusters. I very occasionally see an issue where the 'ISALIVE:0' check will fail for a virtual machine (via the vmclusres.dll library) and the Resource Hosting Subsystem will terminate, bringing down multiple machines with it.

There are multiple pieces of contradictory information on the internet about what is actually happening here. Some sources suggest that fail-over clustering will attempt to isolate the resource which first failed the health check into its own process (suggesting this protects the other resources running under the same RHS).

This categorically does not happen on a fully patched Windows 2016 Hyper-V cluster. What appears to happen is that the failing RHS is terminated, killing any and all compute that is running under the same process. The logs make mention of the offending virtual machine resource being isolated, but I can actually see no evidence of this occurring (in the properties of that resource), but even if this does happen, the default config has still resulted in one resource effectively causing an outage.

I can sort of add credence to this summary of behavior but forcing the resource to run in their own separate monitors myself. If I do this on a lab I've stood up:

Get-ClusterResource -Name "*Virtual Machine blah*"

foreach ($resource in $cluster_resources) {$resource.SeparateMonitor}

... I can see they are all using the default setting, which is to not run in a separate monitor. Fine.

If i set them all to run in their own monitor:

foreach ($resource in $cluster_resources) {$resource.SeparateMonitor = 1}

...and count the RHS processes there is no difference. As you would kinda expect, if I now restart the machines I suddenly have lots of RHS processes popping up, one for each VM.

So this suggests that resource cannot magically flip between RHS parent processes while they are running\switched on, so the out-of-the-box configuration can indeed bring down a whole node when there is a problem with a single resource. Could anybody anywhere tell me if I am right here?

Also, trying to go back to the why behind the original problem. Does anyone know where I can get information about what the ISALIVE check for the vmclusres.dll library is actually doing? There is no information anywhere about what kind of check failed, if it is a VM state check, some kind of IC communication check etc The VM didn't dump inside the guest, it just 'failed' and caused an outage, which is kinda scary. I know from some research that the ISALIVE check is the five-minute check which is supposed to be the more in-depth check of the two that are run, but I can find no documentation saying what it is actually checking, therefore I have no way of working backward.

Thank you.

If anyone can provide a quick answer to this, i would be greatly appreciated.

I have a server 2012 hyper v core.

I was managing it remotely from another location, converting a vhd to a vhdx and it failed due to loss of connection.

When i remoted to the host, the VM will not start, and has no error details besides failed state.

I am able to start the VM just fine if i remove the vhd in question.

I was going to restore a VHD from backup, and noticed i could not delete the VHD i was converting as it stated still being used by another process. I noticed a few delays loading the vhd in file explore also indicating its in use.

Main question i have is, could the host still be in the process of converting the vhd to vhdx and i just have no graphical representation of its current status ? 

If so, could i potentially wait a few more hours and then try to delete the vhd or try to start the vm indicating the conversion finished ?

Hello!

After upgrading to Hyper-V 2016, I faced the fact that now there is no possibility to connect to the server via the RDP, the connection simply gives an error on timeout. In the log on the server, I see events with numbers 4624, 4776, 4672, 4634 ..
It seems that when I try to connect, I successfully pass authorization, but then something goes wrong.
I have already tried to allow rules for the RDP in the firewall and then allow all connections.
There were no problems with the RDP before the update.

Please, what are the possible causes and solutions to the problem?

I have a technical questions regarding AVMA.

I have a valid Windows 2016 Server Datacenter license and a Hyper-V role installed. During the installation of a virtualised Windows server 2016 edition, I should provide any of the AVMA keys available in Microsoft docs as a serial number, and the machine should activate without any problems? Windows Server 2016 Datacenter key is a MAK one.

I've looked a couple of posts, but I can't seem to find the issue to my satisfaction.  I've already wasted a week on this and I don't feel like wasting one more second.  Hopefully, someone here can help.

The environment:

Host machine: Windows 10 v1809 64 bit, AMD Ryzen 3, 2200G (3.5GHz) 16.0GB RAM (14.9 usable)

Guest Machine 1: (This machine is fine) CentOS v7 64-bit.  Assign 127GB VHD dynamic, a second 64GB VHD dynamic, and an optical drive, 4GB RAM

Guest Machine 2: (The Problem Child!) Windows 10 v170? (v1803 won't install), Assigned 127 GB VHD dynamic, an optical drive, 4GB RAM.  I have manually installed the runtime version of .Net Framework 4.7.2, which is required for the software I'm working on.

The problem: I install the initial software, change all text files I need to change, and generally get the system into a condition that I want to be the "starting point" setup.  Over the course of this process, I make about 4 to 6 snapshots and restart 2 or 3 times without an issue.  After I'm in my starting point state, regardless of what i do, the next time I make even the most insignificant change and choose "Shutdown" or "Restart", be it from the Hyper V guest menu, or from inside the operating system's screen, this guest WILL NOT restart!  The system shows as running in the Hyper V manager, and even shows the login screen in the miniature screen capture.  Attempting to run the interactive client, which is absolutely necessary, starts an infinite process.

Initially, I see the Hyper V screen, with the Windows 10 "chase dots" at the bottom of the screen.  The screen then changes to the "Connecting to {guest name}" screen, waits about 30 seconds, then switches to what should be the login screen for Windows 10...but it never appears.  The screen resizes to the correct size, but the screen never appears.  After about 45 seconds of a black screen, the interactive screen for the VM returns to the "Connecting to {guest name}" screen, where the process start again.  It goes through this four or five times, burning about 6 minutes, then, on the last "Connecting to {guest name}" screen, a pop-up dialog appears reading that something couldn't be started, with the options to "retry" or "cancel".

Choosing "Retry" starts another 6 minute cycle of screen switching, resulting in the same dialog box popping up.  Choosing "Cancel" dismisses the interactive screen, even though Hyper V Manager still shows the guest is running and shows the login screen as the active screen.

I've tried most of the suggestions on a similar issue posted about two weeks ago, (the system won't let me post the link).  I do not have the ability to repartition my main hard drive.  I've also uninstalled and reinstalled Hyper V twice.  I found it interesting that, both times when Hyper V installation was complete, it was able to find the previous VMs, even though those weren't installed in the standard directory; I placed the VMs and their respective VHDs on a separate drive, because I didn't want to hunt around in "C:\Program Files" to find them.  I just thought it was odd that, after uninstalling and reinstalling, the system could find my previous setup, checkpoints and hard drives.  Maybe that's part of it, maybe not.

Additionally, I've tried recreating the guest with the image, and also recreating the hard drives, and both of those things together: all have the same result: On the fifth or sixth snapshot, even the most insignificant change, the one that has me frustrated enough to finally post is I changed one, four letter word in a text file to a five letter word, causes the infinite boot process I described above.  I may not even be a boot problem, but a graphics problem, since Hyper V manager is showing the system is booted to the login screen, and that the system is running.  It's just trying to run the interactive session that's an issue.

Other colleagues on my team have this entire system running fine.  No one here has any clue about my problem, let alone how to solve it.  I keep getting "that's really weird" as a response from pretty much everyone here.  I've followed the exact steps they've told me, and I seem to be the only person with this issue.

I'm hoping I'm not really alone in the world.  Because that's a really scary place to be.

Any and all reasonable responses will be considered.  I'm going to reinstall Hyper-V now.  I tried VirtualBox, but that system...their support forums leaves MUCH to be desired.

***EDIT***

The pop-up diaglog reads: "The session was disconnected.  If you want to continue, try to connect again.  If the problem persists, contact your system administrator.  Would you like to try to reconnect? Reconnect/Exit"

Hi,

I installed Hyper-V Server on a spare box to use it as a test environment. All installation and config went smoothly, I turned off server firewall and joined my domain to simplify testing process. I'm able to do most remote management tasks: I can RPC to the server, Computer Management, Hyper-V Management, the whole enchilada. But I seem to be having some issues managing server disks. In Hyper-V Management when I try to create a new VM, it only allows me to save it to the boot (C:) drive (there are 2 other disks in the server that are nowhere to be found). Then I remote to the server with Computer Management to verify the disks, "System Tools" are working fine but when I click on "Disk Management" it fails with error "Virtual Disk Manager: The RPC Server is unavailable". Using Server Manager from my AD, however, I can see all server drives.

In summary, all I need is to be able to select drives other than C: while creating VMs using a Remote Hyper-V Manager. What am I missing?... Thanks!

Hi.

I'm wanted to setup some vlans in a vm inside HyperV, this machine has 2 NICs, I want to allow just 1 NIC for vlans, but onces I run the command:

Set-VMNetworkAdapterVlan

It applies  to both interfaces, I still wondering if HyperV allow me to run the command to a specific vm interface.

Both are External.

Running HyperV 2012.

Thanks.

Hi,

I'm trying to enable Hyper-V on Windows 10 v1809 Pro N, but I'm running into what looks like a pretty nasty bug.  I'm following this document: https://docs.microsoft.com/en-us/virtualization/hyper-v-on-windows/quick-start/enable-hyper-v

When I enable Hyper-V using this one-liner,

Enable-WindowsOptionalFeature -Online -FeatureNameMicrosoft-Hyper-V -All

it asks to reboot (as expected) but my machine goes into reboot-from-hell -- automatic repair, etc.  I was able to boot into safe mode, disable the Hyper-V feature again, but then it wouldn't even boot into safe mode!  I had to re-install Windows.

If I run the above Powershell, but instead of rebooting, I go into the GUI "Turn Windows Features on or off" (as instructed in the linked document).  I see that the Hyper-V check-box is checked (with a check mark).  Of course, it still crashes when I reboot.

When I don't run that Powershell command, but I use the GUI to enable Hyper-V, the Hyper-V check-box is only partially checked.  Clicking that check-box multiple times will toggle between unchecked and partially checked.  I cannot make it fully checked (with a check mark).  Rebooting with it partially checked works correctly. It restarts fine and Hyper-V is enabled.

I've tried various variations on the Powershell script, but I can't seem to find a script that leaves it in the same state as when you enable it using the GUI.  Seems like a bug to me, and not just a Powershell bug -- the GUI looks weird too.

Explanatory image here: https://chris.heathens.co.nz/images/win10proNbug.png