vim.event.NotEnoughResourcesToStartVmEvent is triggered , it is only for 1 VM, not for other VMs on the same Esxi host

Products

• VMware vCenter Server

Issue/Introduction

End user found the following alarm about a VM:

 

vim.event.NotEnoughResourcesToStartVmEvent

 

Insufficient resources to fail over this virtual machine. vSphere HA will retry the fail over when enough resources are available

 

It looks like a HA failover warning, but this warning is found only for 1 VM,    the other VMs on the same Esxi host don't trigger this alarm.

Environment

vCenter 8.0.x

Cause

This is not a typical host failure scenario.  The Esxi host is running.

 

When we check the log files on this Esxi host, there are many power-on resets message found in vmkernel.log, it means that something is going wrong on storage side. It can be caused by a very busy storage array  etc.

So the Esxi host is not failing, and it is not unresponsive.  The abnormal status is caused by the storage access.

And HA is not trying to failing over the VMs, it need to find a placement for the VMs before failover. And the following message is returned for the placement:

fdm.log

YYYY-MM-DDThh:mm:ss.###Z Db(167) Fdm[4513715]: [Originator@6876 sub=Placement opID=57d8d889] Vm /vmfs/volumes/#######-########-####-##########/############/############.vmx failed placement with fault [N3Vim5Fault21NoActiveHostInClusterE:0x00000023b4d7f630]

YYYY-MM-DDThh:mm:ss.###Z Db(167) Fdm[4513625]: [Originator@6876 sub=FDM opID=WorkQueue-1016e731] New event: EventEx=com.vmware.vc.HA.ClusterFailoverActionInitiatedEvent vm= host= tag=host-######:732390281:1

YYYY-MM-DDThh:mm:ss.###Z Db(167) Fdm[4513458]: [Originator@6876 sub=Placement opID=placementService.cpp:151-4fa44a37] [hdcs-ha] Final Vms sent to placement engine. ESX Agent Vms = 0.HDCS Agent VMs = 0, VMs = 1.

YYYY-MM-DDThh:mm:ss.###Z Db(167) Fdm[4513715]: [Originator@6876 sub=Placement opID=57d8d889] Reset the placement status of vm /vmfs/volumes/#######-########-####-##########/############/############.vmx to 0, vim.fault.NoActiveHostInCluster (vim.fault.NoActiveHostInCluster)

YYYY-MM-DDThh:mm:ss.###Z Db(167) Fdm[4513708]: [Originator@6876 sub=FDM opID=WorkQueue-72470507] New event: Event=vim.event.NotEnoughResourcesToStartVmEvent vm=/vmfs/volumes/#######-########-####-##########/############/############.vmx host=host-###### tag=host-######:732390281:2

Resolution

Contact your storage vendor to investigate the storage access issue.