Hyperflex Storage Controller VM wll not power on.

[silviomanziano]

Hyperflex storage controller VM will not power on.  Getting an error Module DevicePowerOn power on failed. Failed to start the virtual machine. Device 201:0.5 is not a passthrough device.  Has anyone ever seen this type of issue before. 

[Mark Elsen]

 - Review these bug reports : https://bst.cloudapps.cisco.com/bugsearch?pf=prdNm&kw=Module%20DevicePowerOn%20power%20on%20failed&bt=custV&sb=anfr

 M.

[silviomanziano]

Unfortunately we tried this python script already from the ESXi server and it did not fix the problem.  

message = 'Device 201:0.5 is not a passthrough device.'
},
(vmodl.LocalizableMessage) {
dynamicType = <unset>,
dynamicProperty = (vmodl.DynamicProperty) [],
key = 'msg.vmx.poweron.failed',
arg = (vmodl.KeyAnyValue) [],
message = 'Failed to start the virtual machine.'
}
],
reason = "Module 'DevicePowerOn' power on failed. "

[Steven Tardy]

Have seen similar issues a few times. Most often after a firmware upgrade renumbers PCI addresses or sometimes with failed hardware.
Does this server have NVMe disks? Just the NVMe cache disk or All-NVMe?
What do other servers show as "Device 201:0.5"? I usually edit another SCVM but do NOT make changes.
If the python script doesn't fix, then I'd check from that ESXi host what `lspci` shows and compare to other ESXi hosts.
Could be that a NVMe disk is dead, but really depends on the output of this server and what other servers shows.

Should probably open a TAC case and have TAC assist with the investigation.