I have 2 HyperV 2012 R2 servers in a datacenter and 1 HyperV 2012 R2 server in a disaster recovery site. Both servers at DC have access to shared storage (fibre based SAN) if needed. Ideally, I would like a given VM to reside on DCVM1 and auto-failover
to DCVM2. However, I would also like the possibility of failover (manual is fine) to DRVM1 should the DC fail. What combo of technologies would I need (failover clustering, HyperV replication, etc) to make this happen. I tried replication, but it only let
me replicate to 1 machine. Clustering seems to remove the replication option.
HyperV 2012 R2 Multiple VM Copies
↧
↧
RHEL 6.5 on Clustered 2008R2
Has anybody had any issues installing RHEL 6.5 x86_64 on 2008R2 Clusters managed by SCVMM 2012R2? Although it would load, it would not complete a first boot if any X windows packages were installed. It would boot if I left these packages off but the Ethernet adapter didn't appear even though the kernel modules appeared to be loaded. I did this on two different clusters managed by two different SCVMMs. I'm sure it's something to do with the kernel's paravirtulized drivers for Hyper-V...
If I downloaded the 6.4 ISOs, they installed fine. I'm yum updating them now to see if they break once they get to 6.5.
Thanks,
JAK
***Update*** They hung in the exact same place as a fresh 6.5 install after the 6.4 to 6.5 update "Starting certmonger: OK"
***Update*** After installing the 6.5 updates and selecting the old kernel version on boot, the system successfully booted.
6.4 Kernel version 2.6.32-358 -successful
6.5 Kernel version 2.6.32-431 - failure
***Here's another update of "downstream" Linux distros with the same issues:
http://www.clearfoundation.com/component/option,com_kunena/Itemid,232/catid,39/func,view/id,58122/
↧
VM Creation failed due to "Security account manager (SAM) or local security authority (LSA) server was in the wrong state''
Hi All,
We have windows 2008 R2 core HyperV 4 node cluster setup, while creating VM cluster showing follwowing error''The security account manager (SAM) or local security authority (LSA) server was in the wrong state to perform the security operation".
Even if we browse the CSV volume through windows explorer the error is same.
Please help..
Thanks in advance..
↧
Hyper-V Virtual Machine Uptime reset to zero after recovery point taken
I am seeing a Virtual Machines Uptime, as reported by the Hyper-V Manager, being reset to zero after a recovery point is taken. Can anybody explain this?
NOTE: Internally the guest VM still shows the correct Uptime. This confirms that the VM has not actually rebooted. Also that the VM is still operational while the recovery point is being taken, hence the Uptime is not being reset due to a VM save.
DPM 2012 R2
Windows Server 2012 R2
↧
Better to virtualize domain controller?
I don't pretend to know much about this other than a very basic understanding, so please bear with me.
We've purchased a new server that we'll be using exclusively for hosting RDP for two clients. Each client will have their own virtual server. We're struggling a bit though trying to decide where to put the domain controller.
There are two things we're considering:
1. The physical server runs ONLY a Hyper V role doing nothing else and we load up a 3rd virtual server that will act as the domain controller for the other 2 virtuals.
or
2. The physical server runs runs the Hyper V AND acts as the domain controller.
The guy loading up the box is suggesting the first option. It seems to me, however, that we're adding another level of "stuff" with the domain controller being a virtual rather. My thinking is that everything has to eventually go out through the operating system on the physical machine. Making the domain controller another virtual rather than the operating system on the physical box, adds another step where things have to go through the virtual DC THEN out to the physical box.
One of the arguments for the 1st scenario is that we can allocate resources to it more easily if the domain controller is virtual. I also like the fact that if the domain controller goes to crap, its just one file that needs to be restored and we're up again. He also brought up the fact that the virtual DC could be restarted (after updates, for instance) without having to shut down the other 2 virtuals. Whereas if the DC is loaded on the physical box then the other two viruals would have to be shut down before a restart of the DC could occur. However, I'm not sure that stuff outweighs any performance loss that might occur going through (what I'm perceiving as) the extra step of a virtual DC.
Any insights would be GREATLY appreciated.
Thank you,
Ken
↧
↧
How Can we force the Hyper-V Replica to use a dedicated network between two hyper-v hosts?
Dear Team,
How can we force Hyper-V Replica to use a dedicated network between two hyper-v hosts?
For live migration, we can choose the desired network in Hyper-V Manager (Use these IP addresses for Live Migration).
I have two 10G adapters teamed together, the virtual switch is created on top as converged network with several vNICs (MangementOS, LiveMigratrion, Backup, etc...)
Each network is set with a specific VLAN to isolate the traffic.
The two Hyper-V hosts are on the same LAN and domain.
Thank you.
Regards,
↧
Hyper-V 2012 Core - Replication failing during setup
So I recently setup a new Hyper-V 2012 Core server and migrated 3 VMs - everything is working properly. I then repurposed my old server and set it up as a Hyper-V 2012 Core server for replication purposes of my 3 VMs. Everything was working properly when I had the servers on the same 192.168.0.x subnet. I then paused replication and moved my server out of state to a remote office connected with a site-to-site VPN connection on a 192.168.3.x subnet. I have changed DNS and can ping, nslookup, etc from/to both servers. When I tried to unpause replication it failed.
This was a few weeks ago, so since then I have removed replication and my VMs from my remote server. When I try to re-establish replication for any of the 3 VMs, it errors out saying it can't connect to the server on port 80. Though I can telnet from my primary server to port 80 on my remote server?
What am I missing or what should I look for as I'd like to get replication working again soon.
↧
Running Windows Server 2012 R2 Virtual MAchine on Host Server running Windows Server 2012 Datacenter
Our host server runs Windows Server 2012 Datacenter, which allows for unlimited licensing on Hyper-V virtual machines. We don't have any budget currently to upgrade this to Windows Server 2012 R2 Datacenter.
Most of the Hyper-V virtual servers run Windows Server 2008 Standard, 2008 R2 Standard , 2012 Standard etc.
Can we create a virtual machine and install Windows 2012 R2 Standard server on this host server?
Thanks.
↧
Licensing question for virtual servers under hyperv server
Hi,
My client is planning to buy a new rack server which is going to be either a Dell 520 or 720.
Since now you can only get windows 2012 essentials which excludes exchange (unless you want to take up the cloud offering) I decided that I install Hyperv 2012 R2 server on the new physical server and create 2 VMs which are going to be:
1)Windows 2012 R2 setup as a DC only
2)Windows 2012 R2 as a member server with Exchange 2013.
I know that Windows 2012 R2 STD can have 2 instances but what about licensing?
Do I have to order 2 server licenses, of for each VM or can I get away with one server license for the 2 VMs?
Previously my client has been running Windows 2003 SBS which had included Windows, exchange and a few other things.
Now with the upgrade to the new server, Microsoft is not offering Windows 2012 SBS but Essentials which does not include Exchange. This makes it much more expensive from a licensing point of view.
Can someone steer me in the right direction?
Thank you
Warm regards
ipnotech
↧
↧
Migration Setup
Hi,
I have an existing Win 2012 Server running hyper-v using a HP iscsi SAN. I am adding a second 2012R2 server to my setup. My plan is to migrate the Guest machines to the 2012R2 server, upgrade the 2012 server and then create a cluster. Can I just connect the iscsi volumes to the new 2012R2 server at the same time as the 2012 server and live migrate them over? Are there any issues with connecting multiple clients not in a cluster to the same iscsi volumes or anything like that?
Thanks
Jason
↧
Hyper-V network adapters - 1 per VM?
Hi,
I have 2 VM's running on 1 host. Do I need 2 virtual NIC adapters and two hardware ethernet cables? Will using 1 VNIC and 1 hardware NIC for two VM's impact performance? They're being used for Citrix?
Also what is the impact by having 2 servers using 1 physical network adapter and two vnics?
Thanks, Matt.
↧
WMI High CPU Usage on Hyper-V VMs - Related to Data Exchange Integration Service
Title pretty much says it all. Some of my VMs have high CPU and moderate usage going to the WMI Integration Service. I have tracked it down to the Data Exchange Integration Service. If I de-select the service under the VM configuration, everything works normally. Has anyone seen anything like this yet?
Thanks, TJ
↧
Hyper-V Host Server RAM Utilization
Hi,
We are running Hyper-V Environment on a Dell Power Edge R720 Server which has 128GB RAM.
Installed Windows Hyper-V Server 2012 R2 on the Host and created 7 VM's.
Total RAM assigned for 7 VM's is 116GB and remaining available is 12GB. But when i am trying to create a new VM on this host,it says you don't have sufficient RAM to create a VM. what could be the reason ?
How much memory does the Host server occupy and what factors does this take into consideration ?
11GB for a Server Core installation of OS ? i am stuck here ?
Thanks .
↧
↧
DPM 2012 R2, Windows 2012 R2, and "stop" VM during backup
I didupgrade
DPM2012SP1to2012
R2.
DMPbackup theserverwith Hyper-Vmachines,windows /linux.
The previous version ofWindows,as didbackupviapartitionshapshotnot stopits operation.
Currentunfortunately itdoes!
Duringbackupclosesall connectionsto the virtual machine,startthecopyandreturnconnections.
Hyper-Valsoshows the beginningofa newmachineto counton timeas if theyheldthe restart (even thoughitwas not).
As everychange /off?
Going back to theprevious version?
BackupGroupI createdagain -does not help.
DMPbackup theserverwith Hyper-Vmachines,windows /linux.
The previous version ofWindows,as didbackupviapartitionshapshotnot stopits operation.
Currentunfortunately itdoes!
Duringbackupclosesall connectionsto the virtual machine,startthecopyandreturnconnections.
Hyper-Valsoshows the beginningofa newmachineto counton timeas if theyheldthe restart (even thoughitwas not).
As everychange /off?
Going back to theprevious version?
BackupGroupI createdagain -does not help.
↧
Child Partion Backup Not available
Hello all, I've got the following setup:
- Windows 2012 running DPM Server 2012 SP1
- Several Windows 2012 Servers running Hyper-V (2012 + 2008SP1 VM Servers)
For a majority of my VM's on the Hyper-V hosts I'm able to back up using Child Partition Snapshot - but I have a single VM server which refuses to be backed up using the Child Partition Snapshot (even though this server is identical - build from a VMM image as the other servers) and only allows Saved State.
The problematic VM server is Windows 2008R2 - the Hyper-V intergration agents are installed and all services running, the disk is not dynamic, the host is running, VHD is a fixed size and it NTFS.
I've tried rebooting the VM, moving to different protection groups without any change to the Saved State backup - does anyone have any ideas?
Regards
Simon
↧
Error Creating Snapshot
While trying to create a snapshot of any of my VM's I am getting an error.
Environment:
Windows 2012 R2 Cluster
Running System Center VMM 2012 R2
Clustered Shared Drives
Any idea what this "No such interface supported (0x80004002)" means or how I fix this?
Error (12700)
VMM cannot complete the host operation on the TUKC-HYPERV-5.corp.bittitan.local server because of the error: Checkpoint operation for 'TUKW-MWProd-106' failed. (Virtual machine ID 53B681FD-8F83-4272-861E-033101EEBB4E)
'TUKW-MWProd-106' could not initiate a checkpoint operation: %%2147778816 (0x80048100). (Virtual machine ID 53B681FD-8F83-4272-861E-033101EEBB4E)
'TUKW-MWProd-106' is unable to reset the virtual hard disk path for C:\ClusterStorage\TUK-HyperV-A-B01\TUKW-MWProd-106-1\VMM-BASE-2012R2.vhdx as a result of failure to create a checkpoint. No such interface supported (0x80004002). (Virtual machine ID 53B681FD-8F83-4272-861E-033101EEBB4E)
Unknown error (0x8000)
Recommended Action
Resolve the host issue and then try the operation again.
Vincent Fournier
↧
Error code 1603 while trying to install Integration Services on an XP Pro SP2 "repaired" P2V VM
This is actually a solution. I'm posting in case someone else runs into this problem.
I was experimenting with a trying to revive a dead PC. It was an old Pentium III running a legacy app that had a hard disk fail. I had a very old Ghost image but recent "data-only" backups.
Since I didn't have a PC to spare, I decided to try to Ghost the old image onto a VHD. I was successful--VHD mount by AppAssure was very helpful.
But booting the VM resulted in a BSOD. No problem I thought. A repair install using the XP Pro CD should get it going. It did. Successfully boots but lacked Integration Services. Able to login as the old local admin account.
Tried to install Integration Services as usual but part way through, after "...framework install" or "ustor...something" and during "guest services", I got a 1603 error code.
After much fussing around I found that deep in the log files in C:\Windows (look for the most recent ones), there was an error trying to create the directory: C:\program files\Hyper-V Integration Services". I was local admin and the folder permissions for C:\Program Files" explicitly had full control for the local admin account so I couldn't understand why there was a problem. Just on a lark, I gave "Everyone" full control over "C:\program files" and afterwards, re-running the Integration Services carried on to finish.
I'm too much of a noob to know what happened but I'm just reporting what I found was the solution.
↧
↧
Application running slowly under Hyper-V (factor of 100 times slower)
I have an application running in a non virtual Windows Server 2008R2/SP1 environment. It processes about 60,000 records from an indexed file (about 20MB in length) in about a second running in that environment. I want to move the application to a Hyper-V virtual server also running Windows Server 2008R2/SP1. When I do so, the application runs, but processing the same file takes almost 2 minutes.
The application is the only thing running in the virtual machine and the virtual machine is the only VM running on the physical box. The VM is configured with a single processor, 2048M of fixed (not dynamic) memory and a fixed size disk of 255GB. There is no network traffic so the single network connection is not used. The physical machine is a quad core with 8GB of memory and two 1TB drives that are less than 1/4 used in each case.
Now for the weird part. The application runs from the command line. If I start it and do nothing else, it takes about 2 minutes to complete its process. But, if I start the application, switch focus to another window on the virtual machine then switch back, the process completes immediately upon switching back to the command line window.
Any thoughts?
↧
Hyper-v performance issue?
Hi everyone. Quick question on some perf issues i've had for quite some time. To be fair i was using esxi 5.x and had similar issues. My vm server is running a first gen intel core series cpu (i7-920) that is overclocked to 3.8ghz. Just so you don't have
to look it up it has 4 cores (8 with hyper-threading) I have 3 vm running on this server. 2 of the VMs pretty much do next to nothing they are just there for when i need to do some testing (one has server 2012 R2 and the other has ubuntu). The main vm that
i'm having issues with has Win 8.1 enterprise. I don't have a lot of things installed on it but i do have plex (when i want stream videos through my house). At least for 1 stream (never tried more) it works just fine. When i remote into the machine and run
some things its so dang slow. For example a fresh install of google chrome will constantly use 50-60% of the cpu. Somtimes it will peg it. Its go so bad i know use firefox most of the time though a lot better the cpu usage for firefox is still not where it
should be. Now i know chrome is turning into a hog but it shouldn't be that bad. I've had this problem for quite some time (with windows 8 too). I think i have two vCPU assigned to this VM, 2 to Server 2012, & 1 to Ubuntu but like i said they hardly do
anything. I have 4GB assigned to Win 8.1, 8GB to Server 2012, & 4 to Ubuntu (the machine has 24GB Total). Any ideas? Is the I7-920 getting so old that i should have these issue?
Edit: I'm running Hyper-V on Server 2012 R2
Edit: I'm running Hyper-V on Server 2012 R2
↧
multihome network adapter loses connection on additional IP addresses after move
I have a VM that has two NICs, one facing inward to the local network and the other NIC facing to a public Internet.
the public nic has 13 of the 14 available public fixed IP address block assigned to it and the VM responds to all of them just fine, that is, until the VM is moved to another server.
when the VM is moved to another host a peculiar thing happens. the VM responds just fine to the main IP address of the NIC, but any packets sent to the other 12 IP addresses configured on that NIC, the packets never make it to the NIC. or at least I can't see them and neither can the VM.
the two hosts that the VM is moved between both have exactly the same type network adapter and they connect to a Cisco router that does routing duties to the public internet and it sits on the 14th IP address of the public fixed IP address block as the default gateway.
if I move the VM back to 1st host, all IP addresses respond, if I move it back to the 2nd host, the VM responds only to the main IP address assigned on the NIC.
I can recover from this by cycling power on the Cisco router! then all IP addresses respond on the host that the VM was on when the Cisco had its power cycled. if the VM is moved to the other host the same issue again.
I guess this works because it forces the router to refresh its internal routing table and MAC to IP to port assignment table.
but there has got to be a better way than cycling power on the router or touching the router or switch in any way.
any suggestions on how to overcome this issue without resorting to restarts and cycling power would be greatly appreciated!
↧
More Pages to Explore .....
