Description of problem:
Installed openstack director on a single baremetal with 1 controller + 2 computes.
Run tempest .
Nova repeated Traceback :
raise exception.InstanceNotFound(instance_id=uuid)
InstanceNotFound: Instance b6f5b3a3-2cef-48e9-b163-d66238bf6058 could not be found.
_update_usage_from_migrations /usr/lib/python2.7/site-packages/nova/compute/resource_tracker.py:786
2015-12-01 23:55:59.159 19222 DEBUG nova.compute.resource_tracker [req-649995ae-8c01-499e-95c7-345635f3e0ee - - - - -] Migration instance not found: Instance 9fb23e05-e4c4-4f79-bac5-907aa8d28349 could not be found.
Traceback (most recent call last):
This message should be printed once when error happended and not flood nova logs
.
Version-Release number of selected component (if applicable):
How reproducible:
Steps to Reproduce:
1. Run tempest nova livemigration tests
2. migration fails and logs are flooded with same error.
Actual results:
Expected results:
Additional info:
This is simply how nova works ... if the instance is deleted while doing an action, then the background tasks try to access the no-longer-existing instance and complain in debug log messages
Seems like correct (if annoying) behaviour.