[integration tests][nova] TimeoutExpired: Timeout of 60 seconds expired waiting for volume to be deleted
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Mirantis OpenStack |
Invalid
|
High
|
Georgy Dyuldin |
Bug Description
Test result:
https:/
Trace:
cls = <mos_tests.
os_conn = <mos_tests.
@classmethod
@pytest.
def volumes(cls, os_conn):
"""Volumes cleanUp"""
volumes = []
yield volumes
for volume in volumes:
> common_
mos_tests/
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
cinder_client = <cinderclient.
volume = <Volume: dbe45541-
def delete_
"""Delete volume and check that it is absent in the list
:param cinder_client: Cinder API client connection point
:param volume: volume
"""
if volume in cinder_
if len(volume.
> waiting_for='volume to be deleted')
E TimeoutExpired: Timeout of 60 seconds expired waiting for volume to be deleted
mos_tests/
Changed in mos: | |
milestone: | 10.0 → 9.0 |
As we can see from http:// cz7776. bud.mirantis. net:8080/ jenkins/ job/9.0_ Neutron_ NovaCeph/ 2/testReport/ mos_tests. nova.nova_ test/TestNovaDe ferredDelete/ test_restore_ deleted_ instance_ 86400__ _842493_ _/ , it's a cleanup issue, not related to the test scenario. build_number: openstack_ version: network- checker| nailgun| packetary| shotgun' : 9.0.0-1. mos6346. noarch -cli-9. 0.0-1.mos282. noarch 9.0.0-1. mos8386. noarch astute- 9.0.0-1. mos746. noarch 9.0.0-1. mos8386. noarch checker- 9.0.0-1. mos72.x86_ 64 9.0.0-1. mos136. noarch -metadata- 9.0.0-1. mos8693. noarch 9.0.0-1. mos8386. noarch mcagents- 9.0.0-1. mos746. noarch ing-scripts- 9.0.0-1. mos8693. noarch fuelclient- 9.0.0-1. mos315. noarch 9.0.0-1. mos270. noarch 0.0-1.mos6346. noarch 9.0.0-1. mos8386. noarch 9.0.0-1. mos6346. noarch 0-9.0.0- 1.mos8386. noarch 9.0.0-1. mos89.noarch 9.0.0-1. mos282. noarch 9.0.0-1. mos2699. noarch 9.0.0-1. mos934. noarch packetary- 9.0.0-1. mos136. noarch 9.0.0-1. mos8693. noarch
Anyway, I've tried to reproduce it on env (Cinder Ceph + Nova Ceph):
[root@nailgun ~]# shotgun2 short-report
cat /etc/fuel_build_id:
398
cat /etc/fuel_
398
cat /etc/fuel_release:
9.0
cat /etc/fuel_
mitaka-9.0
rpm -qa | egrep 'fuel|astute|
fuel-release-
fuel-bootstrap
fuel-migrate-
rubygem-
fuel-misc-
network-
fuel-mirror-
fuel-openstack
fuel-notify-
nailgun-
fuel-provision
python-
fuelmenu-
fuel-9.
fuel-utils-
fuel-setup-
fuel-library9.
shotgun-
fuel-agent-
fuel-ui-
fuel-ostf-
python-
fuel-nailgun-
and the results are: http:// paste.openstack .org/show/ 506315/ , the volume has been detached and deleted successfully.
Also, I have a question concerning: volume( cinder_ client, volume): client. volumes. list():
volume. get() attachments) > 0:
volume. detach( )
cinder_ client. volumes. delete( volume)
volume_ id = volume.id
wait( lambda: not is_volume_ exists( cinder_ client, volume_id),
timeout_ seconds= 60,
def delete_
"""Delete volume and check that it is absent in the list
:param cinder_client: Cinder API client connection point
:param volume: volume
"""
if volume in cinder_
if len(volume.
> waiting_for='volume to be deleted')
we can see that detach() method is called within cinder-client, but in fact it's Nova's action (nova-attach, nova-detach), and I'm not sure now that it's cinder-client's one as well. Is the nova-client enabled earlier for these operations?