Quantcast
Channel: Hyper-V forum
Viewing all 19461 articles
Browse latest View live

How to make a Hyper-V 2012 cluster perform

$
0
0

Hello,

I am in the final stages of selecting components for Hyper-V 2012 cluster and SAN, I am interested in people's opinions on what the best tweaks for improved performance.

SAN will be 8G FC (eg 3PAR 7400 / Compellent SC8000 / NetApp VNX 5500) with 30-40 Tier-1 disks, 20-30 Tier-2 disks, possible some SSD.

Servers will be 2 CPU (eg E5-2680), 256GB Memory (16x16 1600Mhz), 3 x 146 15K in RAID-1 + spare, 2 x single port 8Gb HBA, 4 x 10GB NIC, 4 x 1GB NIC

Introducing 10Gb switching for east-west traffic that will most likely be connected to 6509-E core using multiple 1Gbps.

Where would you spend money if you wanted to increase performance ?

SSD for the SAN ?

10Gbps from rack switching to 6509 core ?

Upgrade CPU to E5-2690 ?

Fusion IO ?

One vendor suggested having SSD for the Hyper-V OS drives.

I do realise I have provided little details on the workload, assume typical small business, 1000 users, 60 guests, file, print, exchange, sql, sharepoint. Backup is on seperate hardware using DAS.

Thanks


Unable to open virtual machines in windows server 2012

$
0
0


Hi

While trying to open any virtual machine for creating any virtual machine I am getting the following error in Windows server 2012 datacenter edition.VMName’ failed to initialize.Could not initialize machine remoting system. Error: ‘Unspecified error’ (0×800070490).Could not find a usable certificate. Error: ‘Unspecified error’ (0×800070490).

For Hyper-V with Windows 2012

VMName’ failed to initialize.Could not initialize machine remoting system. Error: ‘Unspecified error’ (0×80004005).Could not find a usable certificate. Error: ‘Unspecified error’ (0×80004005).


esxi 4.1 to hyperV server 2012

$
0
0

hello,

I have migrated 6 servers esxi to hyperV server and it's OK. The server look fine but one of them is slow.

it's the server with hyperfile and postgre ! it's 3x more longuer than the old server.

the new server is an IBM x3650 M4 with 80 Go de RAM and 5 SAS disk 500 Go.

the VM have 32 Go of RAM and 400 Go, 2008 R2

Thanks for you help.

Windows server 2012 failover cluster error: Cluster resource 'Virtual Machine Configuration ... of type 'Virtual machine configuration in clustered role ... failed.

$
0
0

I have two windows 2012 host server that are clustered using windows failover cluster feature. Each server is hosting four VMs. When migrating from Host2 to Host1, the migration failed with the following error:


Cluster resource 'Virtual Machine Configuration SCPCSQLSRV01' of type 'Virtual Machine Configuration' in clustered role 'SCPCSQLSRV01' failed. The error code was '0x569' ('Logon failure: the user has not been granted the requested logon type at this computer.').

When this happens, the VM that I was migrating can no longer be started even on the original host. The only remedy is to restart the host server.

Any suggestion on resolving this problem?


Thanks

Ikad

Is it possible to create non-persistent virtual machines on HyperV?

$
0
0

Hi All

Is it possible configure VM on Hyper V that it started every time from VM template and doesn't save any changes after shutdown?

Windows Server 2012 Hyper-V - Query total memory on guest with dynamic memory

$
0
0
Is there a way to know, from the guest VM, the maximum memory configured for a VM in Windows Server 2012 Hyper-V?

Paulo Morgado

Hyper-V CPU Compatibility for Clusters Server 2012

$
0
0
For Hyper-V and Live Migration purposes, are X5677, L5640, E5649, and E5620 CPUs all considered in the same family where they can be live migrated to and from within a cluster without the need to enable the "CPU compatibility" option?  I am assuming since they are all 56XX based processors they share all of the same features.  If that is not accurate is there any public documentation that can provide some insight as to what CPUs share the same features and can be migrated across?

Configuring NIC's in a virtual environment

$
0
0

Hi everyone,

I hope someone can help me in this question.

My scenario is a replica based scenario with WORKGROUP and is as follows:

First Physical Server: W2KHV01

       - VM1: First Domain Controller (domain: test.local)

       - VM2: Microsoft Exchange 2010 (domain: test.local)

Second Physical Server as Replica Server: W2KHV02

       - Enough storage to keep the replica virtual machines.

And my question is: to configure the networking of the first server W2K12HV01 and its virtual machines, wich is the best way to accomplish this? I've thought to assign 1 external NIC of the virtual machine to 1 physical NIC, and another external NIC to the another physical NIC (1 NIC = 1 VM).

The server has 2 NIC's, and I think the best way to have a good performance is assign each adapter to a virtual external adapter.

Who can tell me if the scenario is ok and if my NIC's idea is good for this implementation? I'm Reading many tutorials about this solution but I want the opinión of someone more.

Thanks in advance.

Marc.


Tecnico Superior de Sistemas


VMBUS Yellow Triangel?(windows 2008 (server)

$
0
0

If i got to my device manager I see at the bottom VMBUS with a yellow triangle if i see what the error is it says:

This device cannot find enough free recourse that it can use if you want to use this device you need to disable one of the other devices on the system

so what is VMBUS actually and is this bad that i have a warning next to it ?

and the most important question how do I Fix this ?



NT Virtual Machine\Virtual Machines overwritten by group policy, have to constantly restart VMMS

$
0
0

Hey Guys,

So my root group policy has "logon as service" defined thus overwriting my hyper-v servers local policy. Because of this I keep having to restart VMMS. So I find this article below:

http://support.microsoft.com/kb/2779204

I installed group policy editor on my hyper-v server and went to add "Virtual Machines" as they request but I cannot find the security group NT Virtual Machine\Virtual Machines 

So how do I add this permission? 

Thanks

Looking for advise....

$
0
0

We currently are running 5 ESXi 4.1 hosts and looking to change products. I have long time been a fan of Citrix products but when deploying a lab environment of Xenserver I noticed that it doesn't support thin provisioning with our fiber SAN.

I've been reading a lot about Hyper-V 2012 and it looks pretty promising and i'm impressed, so I think I will be setting up a lab with it tomorrow.

Anyways here are some of my questions (keep in mind i'm a Hyper-V noob lol):

  • We use about 25 VLANs and NIC teaming. From what I have read is the new Hyper-V supports NIC teaming (actually without 3rd party software) and supports multiple VLAN IDs on the NIC team. Can anyone verify?
  • I want to save as much memory as possible and I've read that the host should use about 500MB-1GB of RAM. Now I know (especially with 2008 R2) that it uses WAY more than that. IS this with deploying server core and just enabling the hyper-v role?
  • It  seems the new Hyper-V does actually support thin provisioning for virtual disks... is there any catches?
  • Is there a good way to migrate VM's from ESXi to Hyper-V? (Like Citrix has the Citrix Conversion Manager to assist in migrating from ESX to Xenserver)

Also was just wanting to see if anyone could share their experiencing with going to Hyper-V 2012 from ESXi or Xenserver.

Error trying to mount the DVD to a network based ISO image in Windows 2008 Hyper-V RC0

$
0
0

Greetings,

When I try to configre the guest OS settings to mount the CD/DVD drive to an ISO image I have setting on a file share, I consistently get an error. The only time I have been able to mount an ISO since upgrading to RC0 (I don't remember if I tested this in the BETA) is when the ISO is locally on the Windows 2008 Hyper-V server.

 

To be specific, I get the following error:

Virtual Machine Connection (title of the window)

Inserting the disk failed.

 

Failed to add device 'Microsoft Virtual CD/DVD disk'.

 

User account does not have sufficient privelage to open attachment '\\remoteserver\share\ISONAME.ISO'.

Error: 'General access denied error'

 

'NEWGUESTOSNAME' failed to add device 'Microsoft Virtual CD/DVD Disk'. (VMID 26C8B66F-5D3F-47F9-9D80-474D0F778522)

 

'NEWGUESTOSNAME': User Account does not have sufficient privelage to open attachement '\\remoteserver\share\ISONAME.ISO'.

Error: 'General access denied error' (0x80070005). (VMID 26C8B66F-5D3F-47F9-9D80-474D0F778522)

 

I am not sure what credentials it is using. It looks almost like it is using the new computer name as credentials or something like that. The account I am logged in as at the time is a Domain Admin and has access to every server and every share. I can browse to the ISO just fine through the Hyper-V Insert Disk function, it just won't take.

 

I have checked the Hyper-V server settings and the Guest OS settings and can't find where to specifcy what account to use to connect to remote ISOs. Anyone have any clue on what I am doing wrong?

Free space on physical disk with VHD drive – best practice.

$
0
0

Just wander if somebody could help me to find answer for question about free space on physical drive (in production environment) which I need to have.

Scenario is as follow:

  1. SQL Server on Hyper-V Windows 2012 Virtual Machine
  2. Fixed System VHD on 7.2 K SAS RAID 10 shared with other virtual servers physical Volume
  3. Fixed Database VHD on dedicated RAID 1 600 GB SAS 10K Volume
  4. Fixed Logs VHD on dedicated 300 GB RAID 1 SAS 10K Volume
  5. Hyper-V remote replication to DR site
  6. Altaro backup of Hyper Machines

Questions:

  1. Can I use whole available space on dedicated volumes for Database and Logs? (600GB and 300GB) or do I need leave some space free? If so, could you advise how much?
  2. Any advice about possible problems with following configuration?

Tank you

Hyper-V 4GB mem missing and what about VP's?

$
0
0

Hi all,

We have installed a new Hyper-V 2012 Server stand-alone, with two virtual cliënt OS (Server 2008 standard).

The server has 2 XEON Quad cores and 16GB internal memory.

Both VM's are installed on a (different) RAID-1 146GB container.

One OS is used for SQL Server 2008 and got an amount of 8GB memory.

The other OS is used for ISS (web applications).

I only can give 4GB memory to the second OS, which includes I've loosen 4GB internal memory to Hyper-V, ...right?

I read about a needed 1GB for Hyper-V, but that seems 4GB now?

And my second question is about Virtual Processors.

I plugged 2 QUAD Core processors in the server, which make it up to 8 cores to work with.

I'm able to choose/assign 64 VP's in the Hyper-V manager.

I read on the internet about an unlimited amount of VP's for server 2012, just limited by the used hardware.

But when I tried to assign more than 8 VP's to the SQL server, Hyper-V manager showed me an error: ... assigned more virtual processors than logical processors available...

So.. How do I have to interpret VP's? I read a lot on the internet about it, but just didn't got a clear answer. 

I read about 8:1 ratio's, 12:1 ratio's.. but in my story the ratio seems to be 1:1, right?

Did I do something wrong? This server is going to be used as one of our main servers, so I want to configure it first-time right if possible.

Could someone help me out with this?

Kind regards,

John

Hyper-V Lab Environment internet access

$
0
0

Hi guys I'm looking to setup a Hyper -V environment containing a domain controller which will have WDS/DHCP/DNS server which is completely private of the Host Domain.

The Host is not on the domain but has a static IP with DNS pointing to our network.

My Virtual DC and clients can communicate together using a "private virtual machine network" however they do not have access to the internet.

I don't want to assign a internal network as that then allows the Virtual DC and client see our HOST Domain. (Can ping the domain name as DNS is on the HOST Server.)

Any one have any ideas how I can route only internet to the vm's without them seeing the host domain or even the domain seeing those Virtual clients?

I understand External network will provide full visibility to the domain

Internal will allow visibility via the HOST Server (which is off the domain) but DNS will allow visibility to the network?

Private is completely private which is ideal but no internet access :(

any ideas please



Cannot migrate vm(s) to new cluster node

$
0
0

Original Cluster

           Node 1 - Server 2008 R2 Datacenter SP1

           Node 2 - Server 2008 R2 Datacenter SP1

           Connected to SAN via dedicated switches

           Clustered Storage configured for failover and operational. Live migrations can be performed.

Added 3rd Node

           During validations there were warnings that the shared storage would need to be made available for the 3rd node. I assumed this meant configuring iSCSi Initiator for each volume for the 3rd node once it was added to the cluster which I did. We use CHAP authentication to connect, and all connections tested successfully with Multiple I/O enabled.

I was able to add the 3rd node with no warning or errors. I was able to Live Migrate a server.

The 2nd server I attempted to Live Migrate failed. I attempted an offline migration but the VM failed to start.

I was able to successfully use Storage Manager to give Node 3 control of the Cluster shared storage volume that the VM resides on.

When I attempt to start the VM with Hyper-V instead of Failover Cluster, it gives me the following error.

The attachment the error refers to is an additional volume which I used autoconfigure to add to Node 3's iSCSi Initiator. Please advise.

Shared nothing migration issues (2012)

$
0
0

I'm checking out some new features of 2012 and have 2 standalone servers (not clustered) attached to their own storage (10Gb iSCSI). I've got a dedicated live migration network between the two servers as well (10Gb). I created a dummy VM on server #1 with a few VHDs. Shared nothing is working but I have some questions about if my setup is correct, based on what's going on.

Issue #1 - The migration seems to take a long time and isn't using my available bandwidth. I have 10Gb everywhere but the data is moving through the systems at only between 800Mb - 1.5Gb throughout the whole process. We have great tier1 san (ssds, 15k sas) for our sandbox, so I know the source and destination storage aren't stressed. The data path looks like it comes off storageA via iSCSI into server1, travels over the migration network to server2, and then is pushed out to storageB via iSCSI.

What is the holdup? Why isn't it moving faster and using more of my 10Gb links?  My only thought is that since shared nothing is using SMB 3.0, maybe SMB is throttling things? 

Issue #2 - Possibly related but I'm not sure. I have multiple VHDs on my VM (one per lun for simplicity).  In perfmon as I watch the migration happen, those VHDs are moving one at a time. The destination machine writes to the first LUN only (with no activity on the others), then upon completion it starts writing to the next lun, and so on.

Is it limited to one disk copy at a time? Can it not do parallel copies during shared nothing migration?

Adding Virtual Machines to a cluster but they don't show in the Roles

$
0
0

Hello,

I'm new to the clustering and I had a stand alone server with 8 Virtuals running on it.  I create a cluster and joined this server to the cluster. I first tried to figure out how to move the vmachines in to cluster and gave up.  So I just deleted them in Hyper-V manager and copied the vhdx to the LUN on my cluster and went to Roles and clicked New Virtual Machine and selected the hard drive and all seem to work as excepted. Then I noticed the Machine showed back up in Hyper-V manager is that normal? So then I tried this same approach on a second server it seemed to work but the Vmachine does not get added to the Roles only the first machine is listed in the Roles on the Cluster. It does show in the Hyper-V manager.  Am I missing something? 

What I have it a two node cluster both running server 2012  with a P2000 G3 storage.

Thanks 

Randy


Failure Moving VM to 2012 Cluster

$
0
0

I am having a problem moving VMs from our standalone 2012 Hyper-V boxes to the cluster.  If I try to do a move from a standalone to a cluster node it goes through the entire move process (very slowly compared to how long it takes to do a 'shared nothing' move beteween standalone hosts) and fails at the end with "Virtual machine migration failed at migration source" and "Failed to establish a connection because the destination host rejected the request: element not found."  The error code is 0x80070490.  This must happen after the migration is almost complete beacuse if I look at the destination folder it has the right size VHDs there.  I have tried migrating the storage to the CSV and also to the local direct attatched storage on the target host.

Strangely enough if I do the move from SCVMM it works and it also automatically sets the VM up as a cluster resource.  I don't mind using SCVMM but it would be nice to know what's tripping me up on the other front.  Any assistance would be most appreciated!

Unable to create filesystem (mkfs.ext4) on large > 2TB GPT virtual disk using Linux VM.

$
0
0

I am unable to create a file system on a large (> 2TB disk) virtual disk for a Linux VM.  I can create the disk, attach it to the VM, partition it with "parted", but I cannot run mkfs.ext4.  Details below.

Hyper-V 2012 Core (w/ all Windows/Microsoft updates as of 4/19).

CentOS 6.4 VM w/ 4 virtual processors, 4GB RAM, and 3 dynamic drives: 

/dev/sda  100GB IDE dynamic vhdx
/dev/sdb  75GB IDE dynamic vhdx
/dev/sdc  10TB SCSI dynamic vhdx

Using parted, created 500GB partition on the 10TB drive (/dev/sdc1). 

(parted) select /dev/sdc
Using /dev/sdc
(parted) print
Model: Msft Virtual Disk (scsi)
Disk /dev/sdc: 11.0TB
Sector size (logical/physical): 512B/4096B
Partition Table: gpt

Number  Start   End    Size   File system  Name                Flags
 1      1049kB  500GB  500GB               production_archive

then run: mkfs.ext4 /dev/sdc1

repeating error on console from mkfs.ext4:

INFO: task mkfs.ext4:2581 blocked for more than 120 seconds
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.

Runaway error in var/log/messages until my /var system filled up - 25G worth of it:

-rw-------. 1 root root 25085329408 Apr 19 23:15 messages

Apr 19 17:39:28 nfs2 kernel: sd 4:0:0:0: [sdc] Sense Key : No Sense [current]
Apr 19 17:39:28 nfs2 kernel: sd 4:0:0:0: [sdc] Add. Sense: No additional sense information
Apr 19 17:39:28 nfs2 kernel: hv_storvsc vmbus_0_13: cmd 0x93 scsi status 0x2 srb status 0x6

Same problem happens when running "mkfs.ext4 -E lazy_itable_init=1 /dev/sdc1"

Viewing all 19461 articles
Browse latest View live




Latest Images