HyperV Host / virtual machines pooched after creating new virtual switch.

Question

Evening. Bit of a bind and I'll try hard to explain - short version.
Two physical hosts (HV01, HV02) each has several VM's on them
We have a Nimble for storage - we use cluster management (I did not set this all up so I'm still learning)
The hosts have several NIC. (Coworker was 80% completed with clustering project - never finished)

On HOST1 I tried to create a virtual switch for a couple VM's that required access to a vlan.
Creating virtual switch did not go well. It created an error, and that's why HOST2 went all stupid. After I was able to get access into Host 2 I noticed the first three NIC's I checked had their properties all changed, which I am pretty sure that created this mess I am in.

Pretty much every VM on Host 2 ended in saved-critical state. (Two VM's with configs on the local drive did come up as expected)
We know where the config files/VHDx are located on the Nimble but Cluster Manager has our Host 2 in a locked/orphaned state for some reason. And I don't know why, so any assistance ideas appreciated.

Terry

Answer 1

Hi,

What is the type of the virtual switch, if we create "External Virtual switch", please select "Allow management operating system to share this network adapter".

https://learn.microsoft.com/en-us/windows-server/virtualization/hyper-v/get-started/create-a-virtual-switch-for-hyper-v-virtual-machines

---------------

If the reply could be of help, please help to accept it as an answer, thanks for your cooperation!
Thanks for your time!
Best Regards,
Anne

Answer 2

"The culprit that delayed getting things going was the Fail over Cluster Manager"

The rule of thumb for managing is to manage from the highest level possible. If you are running just Hyper-V, you manage all aspects from the Hyper-V manager. If you cluster Hyper-V, you manage everything you can from the cluster manager, using Hyper-V manager only when cluster manager is not able to do something. If some actions which can be performed from cluster manager but are instead performed from Hyper-V manager, you can end up with a situation like you had where it takes some time for the cluster manager to become aware of the change. Sometimes it goes through just fine, but other times it make take a while. Hence my rule of thumb to manage from highest to lowest.