yahoo-eng-team team mailing list archive

[Sean Dague <sean@xxxxxxxxx>]

** Changed in: nova
       Status: New => Won't Fix

-- 
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/1677892

Title:
  nova scheduler_default_filter ComputeCapabilities filter breaks other
  filters

Status in OpenStack Compute (nova):
  Won't Fix

Bug description:
  im using a Fuel deployed Liberty environment. I have a "test team"
  project and a "dev team" project.

  I set up an aggregate with metadata "devhardware = true". and put some
  older hypervisors in the aggregate. (this is so the dev team will
  create instances on the older hardware and the test team will get the
  newest hardware to test our cloud product on).

  I created a flavor that only the dev team will use that also has set:
  aggregate_instance_extra_specs: devhardware = true

  I added
  AggregateInstanceExtraSpecsFilter,AggregateMultiTenancyIsolation to
  the scheduler_default_filters of nova.conf

  I could not create an instance using that flavor. Creation would fail:

  2017-03-30 16:51:07.670 7773 WARNING nova.scheduler.utils [req-f3e4c44e-2edd-4da0-98ee-265628f2c5c8 e581f58b4ab441f2bb61d3ec9c3bf735 bca6cc5337f44bd089ab4490124b3cff - - -] Failed to compute_task_build_instances: No valid host was found. There are not enough hosts available.
  Traceback (most recent call last):

    File "/usr/lib/python2.7/dist-packages/oslo_messaging/rpc/server.py", line 142, in inner
      return func(*args, **kwargs)

    File "/usr/lib/python2.7/dist-packages/nova/scheduler/manager.py", line 84, in select_destinations
      filter_properties)

    File "/usr/lib/python2.7/dist-packages/nova/scheduler/filter_scheduler.py", line 90, in select_destinations
      raise exception.NoValidHost(reason=reason)

  NoValidHost: No valid host was found. There are not enough hosts
  available.

  2017-03-30 16:51:07.671 7773 WARNING nova.scheduler.utils [req-
  f3e4c44e-2edd-4da0-98ee-265628f2c5c8 e581f58b4ab441f2bb61d3ec9c3bf735
  bca6cc5337f44bd089ab4490124b3cff - - -] [instance:
  b430ff72-4d52-4768-ae75-8ddd06b31337] Setting instance to ERROR state.

  But once I removed ComputeCapabilitiesFilter from
  scheduler_default_filters on nova.conf (which seems was there by
  default) I could create instances with that flavor!! (and they
  correctly were only created on the hypervisors in the aggregate)

  scheduler_default_filters=AggregateInstanceExtraSpecsFilter,AggregateMultiTenancyIsolation,RetryFilter,AvailabilityZoneFilter,RamFilter,CoreFilter,DiskFilter,ComputeFilter,ImagePropertiesFilter,ServerGroupAntiAffinityFilter,ServerGroupAffinityFilter

  #scheduler_default_filters=AggregateInstanceExtraSpecsFilter,AggregateMultiTenancyIsolation,RetryFilter,AvailabilityZoneFilter,RamFilter,CoreFilter,DiskFilter,ComputeFilter,ComputeCapabilitiesFilter,ImagePropertiesFilter,ServerGroupAntiAffinityFilter,ServerGroupAffinityFilter

  seems like a bug, but what do I know... thanks!

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1677892/+subscriptions