Two node 2012 R2 cluster, running on two identical Dell PE R720.
Live migration from Server B --> Server A happens in a flash 3% --> 52% --> 60% --> done
Live migration from Server A --> Server B takes rather long 3% --> 52% --> than it trickles down by 1% for minutes
Any idea what to try to do with it, so the direction of the Live Migration behaves in same quick way (as it can obviously)
Seb
Hello,
I need help with running out of room issue and also understanding with checkpoints. I am currently running Hyper-V Bear and hosting 2 VM's. The host disk is running out of space so the VM's keep pausing-critical which I get. Right Now I am trying to understand a few things. The VM I am currently looking at has 3 checkpoints, the parent and then 2 others.
The parent is a win 7 build the first checkpoint, The second checkpoint is the win 10 upgrade, The third is the checkpoint with all the updates.
The disk is dynamic not fixed. the vhdx file is 87gb and the checkpoints avhdx files: CHKP1-100gb, CHKP2-165gb(added space to the c drive inside the vm) and CHKP-3 is 30GB
1. If I merge the Checkpoints will it free up space by over writing or will it just combine.
2. I try to remove the first checkpoint but then the VM did not work. I am pretty sure it's because it part of the tree but I taught that by creating a new checkpoint would carry over all info so the previous checkpoint would only be needed for rollbacks. Was I wrong.
3. The last checkpoint is the state I want to keep so how do I do so and get rid of the others without affecting the VM itself.
4. I am able to add more space to the disk but wanted to learn how to also clean up VM. I also needed to understand how the space works because there is not alot of data stored on the VM. The C drive is about 165 GB now and I am only using about 87gb so where is all the data coming from that is eating up the host disk?
5. Also in regards to a recycle bin, when I delete a fill is it stored somewhere that I can't see that may be eating up my disk space?
Thanks for any insight that can be provided....
Dears,
i am having a a cluster with 6 nodes, all are Windows 2012 R2 Core. the cluster is managed using SCVMM 2012 R2.
i am trying to remove the cluster from VMM but i am getting error 2606 "Unable to perform the job because one or more of the selected objects are locked by another job."
i went through the MS article https://support.microsoft.com/en-us/kb/2795040 but finally it shows the all tasks that locks objects.
if i deleted these entries what would happen?, or what is the best practice for removing the cluster from VMM?
assistance is highly appreciated.
Samer F. Mustafa Microsoft Technical Team Leader. sf_mustafa@hotmail.com
Hello good people of Technet!
I'm having a problem with my Hyper-V Failover cluster and I was hoping someone could help me out...
Please bear with me, as I have had to figure Hyper-V out for myself mostly, I will happily provide any missing information.
To start off, I have set up the following Failover cluster:
Operating System: Hyper-V Server 2012r2 Core edition, Note: Not Server 2012r2, but Hyper-V Server 2012r2.
Number of nodes in the cluster: 2
Number of physical NICs in use on each host: 8
Storage: Dell Equallogic SAN
Storage access: iSCSI
Port Configuration: 2x iSCSI in a NIC Team
4x Virtual switch, each for a different IP range, not shared with the host OS
1x Management NIC, connecting straight into the main LAN
1x Heartbeat NIC
Between the hosts and the SAN are two switches which we wish to separate from the LAN to isolate the traffic on the cluster and traffic from the cluster to the SAN from the main LAN for security reasons and to comply with Microsoft best practices.
The aim is to make the Hyper-V nodes only accessible via the Management NICs from the main LAN.
The following NICs have connections into the LAN:
1x Management NIC on each server
1x Virtual switch providing LAN access for guest VMs on each server, switch is not shared with Management OS.
All other NICs are connected into the switches, and the switches themselves are currently connected to the LAN by a single connection each.
The problem: When the LAN network cable is pulled from the switches in an attempt to isolate traffic, the Hyper-V guest interface for VMs becomes unbearably slow, as if it's having severe network routing issues (a guess). Also, the traffic from one particular
DMZ subnet (possibly all, but this one's the most problematic) will not route properly to another machine on the same subnet, and communication is lost for these particular guest VMs.
Would anyone have any ideas how I might be able to fix this?
I do apologise if this question seems oddly phrased and vague, I'm fairly new to Hyper-V and as mentioned earlier, I would be happy to provide any additional information.
Many thanks,
Chris Stendall
IT Engineer
Hi,
So I'm running Server 2012 R2 with only Hyper-V. Its installed on a 512Gb 6 gb/s hdd and runs smoothly nothing else installed on that drive. I then have 4 PCIe SSD cards in RAID5 where I install all my guest operating systems on. When I do a read/write test on the server I get about 1000mb/s write and 4000mb/s read. All is good and I'm happy until I do a read write test on the guests. For some reason it only writes 400mb/s and reads at 2000mb/s. It seems as if I have lost more than half the speed and I cant figure out where I'm going wrong, any help would be much appreciated. I've attached some screen shots just in case.
Hi,
We have 2 server with 2 physical nics (pNics) each. Both pNics areTeamed. I have configured converged network with vNics (Management,LiveMigration,HeartBeat) on each server, but it does not work correctly. I get ping reply only on 2 networks (Management, LiveMigration). If disable one of Teamed physical nics (pNic) ping reply start on different 2 networks, like (HeartBeat, LiveMigration). I am not getting all 3 network working on 2 pNics team, what could be wrong?
Network config:
1. Router ports were configured with trunk by network team.
2. Installed Hyper-v and Cluster roles.
3. Configured 2 NICs for network Team (Team1) using GUI. NOT added any VLANs to the team.
4. Run these PowerShell commands:
#1. New-VMSwitch -Name VSwitch -NetAdapterName Team1 -AllowManagementOS $False -MinimumBandwidthMode Weight #2. Add-VMNetworkAdapter -ManagementOS -Name "Management" -SwitchName "VSwitch" Add-VMNetworkAdapter -ManagementOS -Name "LiveMigration" -SwitchName "VSwitch" Add-VMNetworkAdapter -ManagementOS -Name "HeartBeat" -SwitchName "VSwitch" #3. Set-VMNetworkAdapterVlan -ManagementOS -VMNetworkAdapterName "HeartBeat" -Access -VlanId 11 Set-VMNetworkAdapterVlan -ManagementOS -VMNetworkAdapterName "LiveMigration" -Access -VlanId 12 #4. Set-VMNetworkAdapter -ManagementOS -Name "LiveMigration" -MinimumBandwidthWeight 20 Set-VMNetworkAdapter -ManagementOS -Name "HeartBeat" -MinimumBandwidthWeight 40 Set-VMNetworkAdapter -ManagementOS -Name "Management" -MinimumBandwidthWeight 5 #5. New-NetIPAddress -InterfaceAlias "vEthernet (Management)" -IPAddress 192.168.0.2-PrefixLength "24" -DefaultGateway 192.168.0.1 Set-DnsClientServerAddress -InterfaceAlias "vEthernet (Management)" -ServerAddresses 192.168.0.100 New-NetIPAddress -InterfaceAlias "vEthernet (HeartBeat)" -IPAddress 10.10.11.2 -PrefixLength "24" New-NetIPAddress -InterfaceAlias "vEthernet (LiveMigration)" -IPAddress 10.10.12.2 -PrefixLength "24"
I have not assign any VLAN ID on Management network to allow untagged data flow.
Thank you
Small Hyper-V deployment that supports 25 internal staff.
One Hyper-V host to run 4 VMs, the other will act as Hyper-V Replica host in another room. Both servers just have local storage.
VMs: SQL Express, File Server, DC1, DC2
The hosts are in 2 different buildings, with the switches using 2x1Gb (LACP).
Should I team/converge four 1GbE on the Hyper-V hosts, or just do one team for the host management/VMs and use 2 un-teamed NICs for Replica via SMB Multichannel? Also, is it okay to use ReFS as the file system for VHDX files?
4 x 1GbE Team
I figured if I teamed all four NICs, I would create 3 virtual adapters and set up QoS by weight:
Management_Network, SMB_Network1, and SMB_Network2 (all virtual)
2 x 1GbE Team
Management_Network (virtual)
SMB_Network1 (physical)
SMB_Network2 (physical)
Hello,
I am new to Hyper-V. I installed Windows Server 2012 R2 Preview on a physical server and installed Hyper-V role on it.
I am trying to install Windows 2008 R2 on Hyper-V as a guest OS.
From Hyper-V manager, I created a VM and supplied 2008 R2 ISO as a boot device.
Windows 2008 R2 ISO is bootable. I have used the same ISO on VMware and it works.
I connected the VM and powered on, but OS instalaltion doesn't start.
I get error " Boot Failed. EFI SCSI Device 1." Below is th screen capture.
How to fix this problem ?
How do I go to Hyper-v VM BIOS settings ?
How do I change the boot sequence for Hyper-V VM ?
Thanks ! Jayawardhane
Hello, we have a Dell compellent san with 55 disks (7k) and another 55 disks for 15k. It auto tiers the data based on load (higher IO loads will move to the faster disks).
Our Hyper-V VM's greatly struggle. I've tried IO benchmarks on many Hyper-V VMs, and the IO score is terrible, even much worse than my single disk desktop computer with 1 5.4k drive. On the same SAN, speed tests and access times are OK on physically attached servers (non-VM). We also have 100+ms disk access time at times on VMs. We are within the limits of our san, according to Dell, we are not overworking it. Also, we did not have these issues with VMware virtual machines.
This leads me to believe that they Hyper-V layer is greatly slowing down IO. We have a 10 server 2012 cluster 2012 (non-r2).
Has anyone ever had decent IO benchmarks running 200+ Hyper-V VMs?
I am posting one of my horrendous benchmarks below. Please do not mention Azure, I do not really want to pay more money to have servers way out on the internet right now (I've heard a little too much sales BS recently) :)
Looking for people with a local hyper-v cluster and san.
Hi,
I am configuring a new File Server. This time we want to use VM.
By using iSCSI, the VM have a d:\drive which is from the HP StoreEasy Storage Server 2012. The d:\drive size is 2TB.
By using robocopy I have transferred files from the old server to the new VM FileServer (d:\drive).
The file transfer is very slow.
But if I transfer files from the old server directly to the HP StoreEasy it will be fast.
I am not sure what to do now
Thanks
Abu.
Hi ,
We are using Windows Server 2012 R2 and created VM's in Hyper-V . We have more than 10 Hypervisors and each Hypervisor have many VM's. Now the question
is if i know VM's name but sometimes i don't know in which Hypervisor thisVM has created. Could you please guide me how can i find out ?
Hi,
I'm about to upgrade my fileserver from 2008R2 to 2012R2. The server is a VM. For seafty I'll like to create a checkpoint first - but how can I do one with only the OS drive and not all the other drives that is connected to the server (fileshare that the server is deploying)
Can anyone tell me this?
/BirchDK
Dear All,
I have installed windows 2008 R2 VM on guest OS windows 2012 i am able to connect to Physical LAN but i need to assign static IP on VM, I have created a VSwitch to assign WAN IP on VM but i am fail to do that I have 4 NICS first NIC with Physical IP and second IP with WAN... ho can i do this..
Regards
Rabbani
RaSa
I also have the same problem.
If I compact a VHD or VHDX the compact runs for one second and ends, and the VM's remain uncompacted.
I am using REFS format on the drive, and the VM's are Windows 7 x86, if that makes any difference.
I have disabled System Restore in VM's, as that was a known problem with 2008 R2.
Can this be reported as a bug in Server 2012 RC.
I have a hyper-V cluster with shared storage that is giving me the following error:
MAC conflict: A port on the virtual switch has the same MAC as one of the underlying team members on Team Nic Microsoft Network
Adapter Multiplexor Driver
there are duplicate MAC addresses but it looks like they are on the SAN and LAN Network (not on the same network). the network cards are teamed/Multiplexor
I have read in some other threads that this is expected behavior?
Name InterfaceDescription ifIndex Status MacAddress LinkSpeedHi guys.
I'm a student in computer science at university college nordjylland, Denmark and im studying Windows Server 2012 r2 hyper-v.
I want to make 2 VLANs in hyper-v for customers virtual servers, so i can assign a VPN to a vLAN, so the customers cant see his other without going out to the public internet and though the main firewall again. How can I do that in Windows Server 2012 r2 only?
I have tried using Hyper-v virtuel switchs with private, internal, external and NIC teaming without luck. Do I need System Center 2012 r2 to make these things?
Here you can see a an draft example for the architecture:
I have two Windows 7 SP1 machines running on the same Server 2012 R2 Hyper-V host (in a failover cluster). They have the latest updates and integration services. The two have the same hardware settings.
However, one of them shows a value for dynamic RAM demand and status of OK, but the other has empty for memory demand and status. I restarted the Win 7 desktop, but it didn't help. Why is it not reporting? I have a feeling it won't expand either if needed.
