Failed to create vgpu cause of IOError
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack Compute (nova) |
Invalid
|
Undecided
|
Unassigned |
Bug Description
Description
===========
I used 'Tesla V100' to create vm with vgpu.
Got error.
Steps to reproduce
==================
* Create flavor with resources:VGPU='1'
* Create vm with CLI `openstack server create --image 27dc8e63-
Expected result
===============
Create successfully
Actual result
=============
Got ERROR
Environment
===========
1. Exact version of OpenStack you are running. See the following
# apt list --installed | grep nova
WARNING: apt does not have a stable CLI interface. Use with caution in scripts.
nova-common/
nova-compute/
nova-compute-
python-
python-
2. Which hypervisor did you use?
Libvirt + KVM
Logs & Configs
==============
2019-07-22 08:12:18,500.500 21346 ERROR nova.virt.
tags: | added: libvirt vgpu |
While I understand your problem, it's not related to a Nova bug. /gridforums. nvidia. com/default/ topic/9541/ general- discussion/ trouble- assigning- vgpu/
See https:/
In general, it's due to the nvidia driver that doesn't support ECC (yet) so you need to call the same : "nvidia-smi -e 0"
Closing the bug here since like I said it's not related to Nova.