Cluster Hanging Hyper-V 2012 and/or 2012R2

A recent client was experiencing issues in their Hyper-V cluster, where we couldn’t migrate virtual machines nor manipulate the cluster on a particular host.  Not all hosts seemed to exhibit this issue, but we did notice in the event logs there were MAC address conflicts.  The Event ID we were getting was 16945, which you can read about here:

http://social.technet.microsoft.com/wiki/contents/articles/31357.hyper-v-mac-address-conflict-event-16945.aspx

Microsoft describes the way in which a team will utilize MAC Addresses, entirely dependent on which teaming solution you use-dynamic, static, etc. here:

https://technet.microsoft.com/en-us/library/mt179273.aspx

 

However, what they don’t say is this traffic might duplicate the MAC addresses associated with your underlying physical NICs, and then cause communication problems with your cluster.

In order to rectify duplicate MAC addressing, you can assign a new MAC to the teamed interface.  See here for my post on locally administered MAC addressing to use “private” MAC ranges on those teams, so that you can minimize the likelihood of having a duplicate:

http://www.iliketodoit.com/2016/03/10/mac-address-management-and-hyper-v-without-system-center/

Special thanks to Joe @ http://www.imvirtus.com/ for researching the event ID associated with the issue.  What we decided to do was come up with one private MAC scheme for the host pool, which gets handed out to the VMs, and another for the NIC teams that allowed us to identify the primary NIC associated with that team.

Hope this helps, and happy virtualizing everyone!

 

This entry was posted in Hyper-V, Microsoft Technologies, Vituralization and tagged , , , , , , , , , , , , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published.