Failover testing of Node in a 3-Node Single Cluster - HA Failover fails for VM

[SysAdmins]

Getting the following on all VMs when doing a simple test of a single node failure:

Reason: Unable to find healthy compatible hosts for the VM

I'm on a newly built Cluster by HPE Pro. Svcs. I have a single cluster w/ 3x Nodes. Each node is on 4.1.0, when I do a simple failover test - unplugging the 4x 1GB links. The host will fail and then the guests will just become disconnected. I reconnect the NICs and the host reconnects, and the guests will eventually reconnect and state that HA failover for the VM failed. 

I'm still trying to fully understand the benefits of HCI compared to other technologies we have, and i'm just not getting the hype... I understand the single storage/compute/network concept but figured in a single cluster setup, that the failover would be pretty straight forward, but outside of this vSphere error, i'm not seeing much fro the UI. Might have to put on some tall boots and dig through logs, but just wondering if anyone can help in the meantime. Thanks!

[Shivam1]

Hi Sysadmin,

The SimpliVity HCI promises storage HA ,which means that the resillency test you did should involve disconnecting the 10GB nics from the switch (Storage/Federation ) ,which will initate the storage IP failover to the peer node (provided the arbiter is configured and is connected with the cluster during this process) ,during the failover period your VM's will remain on the existing compute ,however will shift its storage to an availaible omnistack controller.

Reagrding vSphere HA failure , it should restart the VM's on other host since one of the host was down ,we need to probe from the logs what failed vSphere HA in your case.

Regards

Shivam