yahoo-eng-team team mailing list archive

[Kristian <oxkipo@xxxxxxxxxxxxxx>]

Public bug reported:

Hi,
I cannot get IOMMU working on my Openstack cloud. Everytime I launch an instance with one gpu I get this error complaining that host doesn't have the requested feature.

Message
Exceeded maximum number of retries. Exceeded max scheduling attempts 10 for instance 769a2108-9a53-4cf5-9055-82411ce5cafd. Last exception: internal error: process exited while connecting to monitor: warning: host doesn't support requested feature: CPUID.0
Code
500
Details
File "/var/lib/kolla/venv/local/lib/python2.7/site-packages/nova/conductor/manager.py", line 480, in build_instances filter_properties, instances[0].uuid) File "/var/lib/kolla/venv/local/lib/python2.7/site-packages/nova/scheduler/utils.py", line 184, in populate_retry raise exception.MaxRetriesExceeded(reason=msg)

Here is the log from nova-compute:
http://paste.openstack.org/show/589527/

And here is a log from nova-scheduler:
http://paste.openstack.org/show/589528/


I'm running Openstack Kolla v3.0.1 on an i7 4790k. My GPU is Nvidia GTX 970 (that I want to passthrough) and my motherboard is Maximus VI Extreme.

IOMMU has worked on this setup in the past on Arch Linux.

** Affects: nova
     Importance: Undecided
         Status: New


** Tags: pci

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Compute (nova).
https://bugs.launchpad.net/bugs/1642419

Title:
  GPU Passthrough isn't working

Status in OpenStack Compute (nova):
  New

Bug description:
  Hi,
  I cannot get IOMMU working on my Openstack cloud. Everytime I launch an instance with one gpu I get this error complaining that host doesn't have the requested feature.

  Message
  Exceeded maximum number of retries. Exceeded max scheduling attempts 10 for instance 769a2108-9a53-4cf5-9055-82411ce5cafd. Last exception: internal error: process exited while connecting to monitor: warning: host doesn't support requested feature: CPUID.0
  Code
  500
  Details
  File "/var/lib/kolla/venv/local/lib/python2.7/site-packages/nova/conductor/manager.py", line 480, in build_instances filter_properties, instances[0].uuid) File "/var/lib/kolla/venv/local/lib/python2.7/site-packages/nova/scheduler/utils.py", line 184, in populate_retry raise exception.MaxRetriesExceeded(reason=msg)

  Here is the log from nova-compute:
  http://paste.openstack.org/show/589527/

  And here is a log from nova-scheduler:
  http://paste.openstack.org/show/589528/

  
  I'm running Openstack Kolla v3.0.1 on an i7 4790k. My GPU is Nvidia GTX 970 (that I want to passthrough) and my motherboard is Maximus VI Extreme.

  IOMMU has worked on this setup in the past on Arch Linux.

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1642419/+subscriptions