Server 2012 R2 Hyper-V cluster nodes BSOD

We have an 11 node Hyper-V cluster running Server 2012 R2 with all latest updates.

We have had an ongoing issue with nodes BSODing or hanging. 

Specs:

Hyper-V Nodes-

128-256GB of RAM per node, depending on generation. 12-32 cores. None of the machines are starving for resources. We have a cluster setup through SC VMM 2012 R2 that optimizes the VMs every 30 minutes. The vhd's for VM's sit in a SMB 3.0 share hosted by a SOFS cluster w/ Storage Spaces on a JBOD. Each Hyper-V host utilizes 2 x10Gb Intel NIC cards (either X520-T2 or X540-T2 depending on node) in a NIC team that is currently set to active-standy

We have tried:

adjusting VMQ settings so each NIC uses different cores

Switching NIC team from Active-Active to Active-Standby.

021715-19625-01.dmp 2/17/2015 3:45:51 PM DRIVER_IRQL_NOT_LESS_OR_EQUAL 0x000000d1 00000000`00000028 00000000`00000002 00000000`00000000 fffff800`25529777 MsLbfoProvider.sys MsLbfoProvider.sys+e777 x64 ntoskrnl.exe+1509a0 C:\Users\wschenk\Desktop\021715-19625-01.dmp 32 15 9600 303,960 2/17/2015 3:48:02 PM

February 18th, 2015 3:15pm

Hi William,

I assume that this is native 2012 R2 teaming? If so what Load Balancing and Teaming Modes doe you have set? Is this team used for any other virtual networks such as csv, live migration etc?

Have you updated the NIC firmware for the cards as well as the drivers?

Is ODX still enabled on the 2012 R2 hosts?
Free Windows Admin Tool Kit Click here and download it now
February 18th, 2015 3:52pm

Hi,

I had the same BSOD today:

DRIVER_IRQL_NOT_LESS_OR_EQUAL (d1)

BugCheck D1, {28, 2, 0, fffff80189b23777} Probably caused by : MsLbfoProvider.sys ( MsLbfoProvider!vmqcGetFirstMappedMNic+f )


I have just two nodes with SAS SAN storage and network with intel 1Gb I350-T4 and Windows NIC teaming also active-standby. NIC firmware and driver are up to date.

My crash occurred during VM reboot. VM is a guest cluster file server node and one of it's virtual NICs lost network connectivity during DPM backup (of files in VM). As I didn't have time to look at the issue I just tired to restart VM and the node crashed :(

Had the same issue with virtual NIC yesterday and also restarted the VM and didn't have a crash...

Cluster node dose not have call current updates (last reboot after November updates). VM has all current updates including new DPM agent version.


  • Edited by Andrzej__P Wednesday, February 18, 2015 5:42 PM nic firmware
February 18th, 2015 5:41pm

Hi William,

Here is an article mentioned the stop error 0x000000d1 when you have a cluster node is running server 2012R2 :

https://support.microsoft.com/kb/2929080?wa=wsignin1.0

Best Regards,

Elton Ji

Free Windows Admin Tool Kit Click here and download it now
March 6th, 2015 11:28pm

This topic is archived. No further replies will be accepted.

Other recent topics Other recent topics