Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

Recently, one of our Hyper-V virtual machines has periodically stopped responding over the network. It seems to be happening every few days, and it occasionally happens up to several times a day. I am by no means a sysadmin, so any direction you guys could provide would be very welcome. I've included everything I know to include below. If you need any additional information, I'll be glad to include it.

  • I can connect through the Hyper-V console.
  • I can't connect to network shares, IIS web apps, using RDP, or using ping.
  • Memory usage seems to be normal (3 of 4 GB)
  • Processor usage seems low.
  • We don't know the exact time the server goes down, but the following error appears consistently around the time it goes down:

    Error 5719, NETLOGON
    This computer was not able to set up as secure session with a domain controller in domain *** due to the following: There are currently no logon servers available to service the logon request. This may lead to authentication problems. Make sure that this computer is connected to the network. If this problem persists, please contact your domain administrator.

EDIT

Here's a few details I missed during the initial post.

  • Both the Hyper-V server and the VM are running Windows Server 2008 Standard.
  • There are 6 VMs on the Hyper-V server, all sharing one external virtual network that runs on a gigabit adapter.
  • The problem seems to be isolated to this one VM.
  • At least one of the other VMs on the server is fairly high load.
share|improve this question
    
Did you create a virtual network for the Server? Also how did you configure the VM when you set it up? What O.S is on the VM ? –  Jemartin Jun 15 '10 at 21:47
    
see if this has anything that applies technet.microsoft.com/en-us/library/ff394763(WS.10).aspx intermitant networking is one of the issues resolved –  tony roth Jun 15 '10 at 22:01
    
@tony roth - I wish that would work. We aren't running R2, just plain old Server 2008. –  Brad Gignac Jun 15 '10 at 23:43
    
the above event is from the guest log correct? Anything in the host log? –  tony roth Jun 16 '10 at 14:25
add comment

7 Answers

up vote 1 down vote accepted

It sounds like there's an issue with the virtual NIC and/or the interacting of a piece of software with the virtual NIC. Here are a few things you can try, but my money is probably on the antivirus/firewall product.

  1. Do all of your VM's have the same antivirus product? Verify that your antivirus/firewall product spcefically supports Server 2008 with Hyper-V, if not, try a different (or temporarily removing if feasible) your anti-virus/firewall product as a test to see if the issue goes away. This was the culprit on our systems, every 24-48 hours a seemingly random VM would lose connectivity until it was rebooted.

  2. Verify that your antivirus product in the management/parent OS has the appropriate folder and process exclusions (MS Article ID: 961804)

  3. Try disabling NIC offload functions such as Large Send Offload & CheckSum Offload in the VM network adapter, they are enabled by default in Windows, but its possible your hardware NIC doesn't support it (or doesn't interact well with Hyper-V) which can cause performance problems and network errors. There are a number of ways to do this, but the quickest for testing purposes is to (within the VM) open up the NIC adapter properties, go to the advanced tab, and disable the offload functions in the list, then reboot the VM. (MS Article ID: 951037) This one seems to be a fairly common issue. You'll probably also see errors on your network switches on the associated ports if it's a problem.

share|improve this answer
add comment

The following article may or may not be related....it was supposedly fixed in 2008 R2, then broken in SP1, then fixed again post-SP1 in this hotfix.

http://support.microsoft.com/kb/2263829

In my experience, this problem still exists, even after applying the hotfix.

To date, I am unable to find a way to resolve this problem. I reckon I have burned more than a week of my time turning off TCP offload and many similar settings; nothing stops the Hyper-V networking stack from failing. I’m not so sure that it is purely network load related as I can make this fail when using ARCserve to backup my Exchange 2010 VM. But it only fails part way through backing up the C: Drive. If I remove the ‘Client Agent for Windows’ and leave the Exchange agent only on the VM, then I can backup the Exchange DB over and over, without a problem. And the rate of data transfer over the (virtual) network when backing up the Exchange DB is much greater than when you backup the C: drive with thousands of little files.

So, this makes me think that it is maybe some sort of file i/o problem with the VHD? Perhaps, an SMB (does ARCserve Backup use SMB?) problem? A combination of both the high file i/o and network load? Maybe a bug in the Hyper-V Integration Services? Something is not right and I can’t believe that there isn’t more noise out there about this. I have 2 servers, at different locations that both experience this problem, albeit the symptoms and recovery are a bit different.

The other server loses the networking stack in the VM, but you have to reboot the host to recover as the VM crashes and becomes unresponsive during the reboot. So, this is more serious insofar that the whole host needs a reboot to correct the failed network on a single VM. This is the symptom reported by jwerwie in the original post.

Mucking around with MAC addresses, TCP Offload settings and so on, would appear to be a colossal waste of time.

share|improve this answer
    
Couple quick additional things that might help: –  superlime Nov 30 '11 at 16:18
add comment

So suddenly one of my VMs became unresponsive. Any other VMs living on the same Hyper-V host could ping the server, but anyone outside the VMHost would get very intermittent response.

It turns out that my co-worker had fired up an old server that we had P2Ved a while ago. The VM still had the same MAC address as the physical server. In my case it ended up being a MAC address issue.

Having said that, when I built my Hyper-V environment I disabled all the offloading features on the server's Broadcom nics that were being used by Hyper-V. I have not had a single network issue until this one.

share|improve this answer
add comment

I would start with basic troubleshooting.

  • Are there any other guests running on the host where the problematic machine is running? If so, are they experiencing any problems, or is it isolated to this one machine?
  • At the time that this machine is unresponsive, is there anything else going on in your environment that utilizes a lot of bandwidth?
  • Do you have the virtual switch configured correctly for that guest machine?
  • Are there any events that show up consistently before your Netlogon error that might point to what is going on?
share|improve this answer
add comment
  1. Is there any chance of an IP address conflict? If the server has a static address, does it overlap with any DHCP pools? Remember that your Windows DHCP pool may not be the only one in your environment, particularly if you have devices like VPN appliances or WLAN controllers.

  2. Are there other VMs sharing the same physical network interface? Do they all have network connectivity when this one doesn't?

share|improve this answer
add comment

I heve this same problem on a virtual machine running on a 3 HV server's cluster.

One of our virtualized win2003SP2 standard servers stops to respond through the network, I enter the server via MS SC VMM or via the Hyper-V console on the node where the VM is and it seems that it lost it's network interface... if I try to go to properties of the network interface I get no window so I'm unable to do Disable/Enable to see if the connection cames back. I have other VM's using the same Virtual Switch on the server and they continue to work without any problem.

To recover the situation I just have to restart the VM or to Migrate it to another Hyper-V node of the cluster.

I saw the System Event Log and I've found several events like the one's below:


Type: Information

Source: netvsc

Category: none

Event ID: 4

The miniport 'Microsoft Virtual Machine Bus Network Adapter' reset.

For more information, see Help and Support Center at

.

or


Type: Warning

Source: netvsc

Category: none

Event ID: 5

The miniport 'Microsoft Virtual Machine Bus Network Adapter' hung.

For more information, see Help and Support Center at

But I really don't understand why they happen.

Best regards, Miguel Branco da Silva Lisboa - Portugal

share|improve this answer
add comment

I have had this exact same problem with some of my VMs. Particularly with Exchange being virtualized. The problem had to do "heavy outgoing traffic" which made our VM lose connection to the network but otherwise looked okay.

I fixed the problem with this patch: http://support.microsoft.com/kb/2263829

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.