We have heard from a few customers about an issue with configuring Network Load Balancing (NLB) on Windows Server 2008 R2 virtual machines. This problem may occur with a Windows Server 2008 R2 virtual machine (guest) on Hyper-V or VMware.
When you configure NLB on a Windows Server 2008 R2 system that is a virtual machine, the system may crash with a blue screen with the error shown below. Another possible symptom is that the guest system may hang.
*** STOP: 0x0000007C (0x0000000000000014, 0x0000000000000002, 0x0000000000000000, 0x0000000000000000)
Researching this issue, we found the issue happens only when some anti-virus products are installed on the Windows Server 2008 R2 virtual machine. If we remove the anti-virus software and then configure NLB, everything works fine.
We have found the following workaround has been successful for customers we’ve worked with on this issue:
Configure NLB on the Windows Server 2008 R2 guest before installing anti-virus software. Anti-virus software may then be installed. If the issue still occurs, please contact Microsoft Support.
- Saravanan N
<p>I have had other problems when configuring Windows 2008 R2 NLB in a VMware ESX environment. My VM's had 2 network interfaces. Once I started configuring NLB the machines just seem to lock up. No reboot or BSOD, but I can't interact anymore. Hard reboot is the way to solve, but NLB never gets up and running. As this setup is situated in a proof of concept environment no antivirus is involved in my case...</p>
<p>Do you you guys have other users reporting this behavior?</p>
<p>I am having the same issue as Thomas. We are running on ESX 3.5 without any AV installed on the VMs. Each VM has 2 network interfaces and they lock up completely.</p>
<p>have you tried unicast or multicast NLB?</p>
<p>Please check the version of NLB.sys. We have the latest version of nlb.sys file and here is the KB</p>
<p><a rel="nofollow" target="_new" href="http://support.microsoft.com/kb/978943">http://support.microsoft.com/kb/978943</a></p>
<p>I am currently workig on a customer case with simlar issue, we are trying out a ESX upgrade and working with vmware. We found that any action through nlb to the NIC hangs the node.</p>
<p>I'm having same issue as above (but a single NIC) -</p>
<p>I have had other problems when configuring Windows 2008 R2 NLB in a VMware ESX environment. My VM's have 2 network interfaces. Once I started configuring NLB the machines just seem to lock up. No reboot or BSOD, but I can't interact anymore. Hard reboot is the way to solve, but NLB never gets up and running. As this setup is situated in a proof of concept environment no antivirus is involved in my case...</p>
<p>This has nothing to do with any antivirus, I have 5 Win 2008 R2 web edition servers load balanced in Vmware ESX server 4 update 1. Randomly they crash whenever opening Network Load Balancing Manager. All 5 servers do randomly, they are all exactly the same with NO anti virus installed at all. Of course there is no event log entry, looks like Microsoft has a NASTY bug in the NLB.</p>
<p>We had exactly the same issue with two Windows Server 2008 R2 virtual machines configured as NLB cluster in VMware ESX environment. Exchange 2010 CAS roles have been installed on top. We have experienced the VM just hangs with no reboot or BSOD. Only resetting VM can bring it back online. We had the latest version of NOD32 Business anti-virus installed on the servers. NOD32 has been installed AFTER NLB configuration took place.</p>
<p>The issue has gone after NOD32 has been completely uninstalled.</p>
<p>We are having the same problem - server 2008 r2 clean install on VMware. Congiguring NLB hangs the server and after rebooting the node shows in the cluster misconfiguration. Strange thing is event log shows it as successfully being completed.</p>
<p>Any resolution for this?</p>
<p>I have exactly the same issue as Thomas above. No antivirus as well. Any solution?</p>
<p>We have several (7 +) Win 2k8 R2 servers and they will randomly just hang. They are virtual guests running on different vSphere 4 U1 hosts across two different clusters. The console freezes, but I can issue a remote shutdown command from another machine and the server will accept it and reboot cleanly. 4 of these machines have NLB, but the other three do not. I've found that if I install NLB through the command line or powershell instead of through Features, that it will install successfully without crashing, but it will still hang up randomly afterward. These machines also do not have antivirus on them.</p>
<p>Same problem. As soon as I add a new NLB cluster (unicast or multicast), the VM is freezing, no blue screen but hard reboot is necessary. Did somebody found a solution of this issue ?</p>