This post has been republished via RSS; it originally appeared at: Core Infrastructure and Security Blog articles.
First published on TECHNET on Feb 19, 2019
Hi everyone, Chuck Timon here to talk to you about an issue my customer contacted me on recently, reporting an issue with Logical Switch deployments on two nodes of a four node Windows Server 2016 Failover Cluster managed by System Center Virtual Machine Manager 2016 (UR6). The error message is shown here - 
The decode for the error – 
I asked the customer to collect a Carmine trace to capture additional details of the failure.
The initial exception - 
The final failure – 
As a test, I asked the customer to deploy a Standard Switch which failed as well. I asked the customer to verify he could create a simple external Hyper-V switch on the hosts and that completed successfully.
I was aware of a problem in SCVMM with Logical Switches 'disappearing' as documented here - System Center Virtual Machine Manager fails to enumerate and manage Logical switch deployed on the host . The customer reviewed the blog and reported that none of the symptoms applied to his environment. We decided to go ahead and re-compile the .mof files mentioned in the blog anyway. After that action completed successfully, the Hosts were refreshed in the SCVMM console. Another attempt was made to deploy a Logical Switch, and the failure was the same. The cluster node was rebooted, and after that, the Logical Switch deployment completed successfully. The process was duplicated on the other node and in the end all Logical Switch deployments were completed, and the cluster was up and running properly. The conclusion I reached was perhaps the WMI service needed to be restarted which, obviously, occurred as a result of the reboot.
Thanks for your attention and I hope this information was useful to you.
Charles Timon, Jr.
Senior, Premiere Field Engineer
Microsoft Corporation