Tempest failure because of LVM bug on precise

Bug #1056213 reported by David Kranz
10
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Cinder
Fix Released
Undecided
Unassigned
tempest
Fix Released
High
Unassigned

Bug Description

THIS IS NOT THE BUG YOU ARE LOOKING FOR

If you got here because of a recheck issue in gerrit, you've found the wrong bug. Please look deeper into the logs on the VOLUMES exercise to figure out why it failed.

******************************************************************************************

There have been some build failures lately that are not consistent involving test_volumes_get and test_volumes_list. These failures can happen on either the volumes or cinder runs. I could not reproduce this locally. Here are some links and the stacktraces:

http://logs.openstack.org/11936/20/check/gate-tempest-devstack-vm/12844/

12:20:59 ERROR: Create a volume without passing metadata, get details, and delete
12:20:59 ----------------------------------------------------------------------
12:20:59 Traceback (most recent call last):
12:20:59 File "/opt/stack/tempest/tempest/tests/volume/test_volumes_get.py", line 99, in test_volume_get_metadata_none
12:20:59 self.client.wait_for_resource_deletion(volume['id'])
12:20:59 File "/opt/stack/tempest/tempest/common/rest_client.py", line 253, in wait_for_resource_deletion
12:20:59 raise exceptions.TimeoutException
12:20:59 TimeoutException: Request timed out
12:20:59
12:20:59 ======================================================================
12:20:59 ERROR: test suite for <class 'tempest.tests.volume.test_volumes_list.VolumesListTest'>
12:20:59 ----------------------------------------------------------------------
12:20:59 Traceback (most recent call last):
12:20:59 File "/usr/lib/python2.7/dist-packages/nose/suite.py", line 208, in run
12:20:59 self.setUp()
12:20:59 File "/usr/lib/python2.7/dist-packages/nose/suite.py", line 291, in setUp
12:20:59 self.setupContext(ancestor)
12:20:59 File "/usr/lib/python2.7/dist-packages/nose/suite.py", line 314, in setupContext
12:20:59 try_run(context, names)
12:20:59 File "/usr/lib/python2.7/dist-packages/nose/util.py", line 478, in try_run
12:20:59 return func()
12:20:59 File "/opt/stack/tempest/tempest/tests/volume/test_volumes_list.py", line 51, in setUpClass
12:20:59 'available')
12:20:59 File "/opt/stack/tempest/tempest/services/volume/json/volumes_client.py", line 113, in wait_for_volume_status
12:20:59 raise exceptions.TimeoutException(message)
12:20:59 TimeoutException: Request timed out
12:20:59 Details: Volume volume86128337728 failed to reach available status within the required time (300.0 s).
12:20:59

http://logs.openstack.org/13618/2/check/gate-tempest-devstack-vm-cinder/2818/

10:57:21 ERROR: CREATE, GET, DELETE Volume
10:57:21 ----------------------------------------------------------------------
10:57:21 Traceback (most recent call last):
10:57:21 File "/opt/stack/tempest/tempest/tests/compute/test_volumes_get.py", line 73, in test_volume_create_get_delete
10:57:21 self.client.wait_for_resource_deletion(volume['id'])
10:57:21 File "/opt/stack/tempest/tempest/common/rest_client.py", line 253, in wait_for_resource_deletion
10:57:21 raise exceptions.TimeoutException
10:57:21 TimeoutException: Request timed out
10:57:21
10:57:21 ======================================================================
10:57:21 ERROR: Create a volume, Get it's details and Delete the volume
10:57:21 ----------------------------------------------------------------------
10:57:21 Traceback (most recent call last):
10:57:21 File "/opt/stack/tempest/tempest/tests/volume/test_volumes_get.py", line 71, in test_volume_create_get_delete
10:57:21 resp, _ = self.client.delete_volume(volume['id'])
10:57:21 File "/opt/stack/tempest/tempest/services/volume/json/volumes_client.py", line 93, in delete_volume
10:57:21 return self.delete("volumes/%s" % str(volume_id))
10:57:21 File "/opt/stack/tempest/tempest/common/rest_client.py", line 162, in delete
10:57:21 return self.request('DELETE', url, headers)
10:57:21 File "/opt/stack/tempest/tempest/common/rest_client.py", line 201, in request
10:57:21 raise exceptions.BadRequest(resp_body)
10:57:21 BadRequest: Bad request
10:57:21 Details: {u'badRequest': {u'message': u'Invalid volume: Volume status must be available or error', u'code': 400}}
10:57:21
10:57:21 ======================================================================
10:57:21 ERROR: Create a volume without passing metadata, get details, and delete
10:57:21 ----------------------------------------------------------------------
10:57:21 Traceback (most recent call last):
10:57:21 File "/opt/stack/tempest/tempest/tests/volume/test_volumes_get.py", line 98, in test_volume_get_metadata_none
10:57:21 resp, _ = self.client.delete_volume(volume['id'])
10:57:21 File "/opt/stack/tempest/tempest/services/volume/json/volumes_client.py", line 93, in delete_volume
10:57:21 return self.delete("volumes/%s" % str(volume_id))
10:57:21 File "/opt/stack/tempest/tempest/common/rest_client.py", line 162, in delete
10:57:21 return self.request('DELETE', url, headers)
10:57:21 File "/opt/stack/tempest/tempest/common/rest_client.py", line 201, in request
10:57:21 raise exceptions.BadRequest(resp_body)
10:57:21 BadRequest: Bad request
10:57:21 Details: {u'badRequest': {u'message': u'Invalid volume: Volume status must be available or error', u'code': 400}}
10:57:21
10:57:21 ======================================================================
10:57:21 ERROR: test suite for <class 'tempest.tests.volume.test_volumes_list.VolumesListTest'>
10:57:21 ----------------------------------------------------------------------
10:57:21 Traceback (most recent call last):
10:57:21 File "/usr/lib/python2.7/dist-packages/nose/suite.py", line 208, in run
10:57:21 self.setUp()
10:57:21 File "/usr/lib/python2.7/dist-packages/nose/suite.py", line 291, in setUp
10:57:21 self.setupContext(ancestor)
10:57:21 File "/usr/lib/python2.7/dist-packages/nose/suite.py", line 314, in setupContext
10:57:21 try_run(context, names)
10:57:21 File "/usr/lib/python2.7/dist-packages/nose/util.py", line 478, in try_run
10:57:21 return func()
10:57:21 File "/opt/stack/tempest/tempest/tests/volume/test_volumes_list.py", line 51, in setUpClass
10:57:21 'available')
10:57:21 File "/opt/stack/tempest/tempest/services/volume/json/volumes_client.py", line 113, in wait_for_volume_status
10:57:21 raise exceptions.TimeoutException(message)
10:57:21 TimeoutException: Request timed out
10:57:21 Details: Volume volume1064768584 failed to reach available status within the required time (300.0 s).
1

Revision history for this message
Thierry Carrez (ttx) wrote :
Revision history for this message
John Griffith (john-griffith) wrote :

As Thierry pointed I believe this is in fact a result of bug 1023755.

Basically anytime the delete on a volume is called right now there's a chance that the zeroing operation (dd) will hang things up on the system. As a result we get these cascaded failures after the first test. I'll update with info from Cannonical if/when available.

Changed in tempest:
importance: Undecided → High
Revision history for this message
Thierry Carrez (ttx) wrote :

Adding openstack-ci as this is #1 problem on the gate right now, training developers to ignore tset failures... and it /might/ not be an openstack or tempest issue, as pointed in comment 1.

Changed in openstack-ci:
importance: Undecided → Critical
importance: Critical → High
Revision history for this message
Thierry Carrez (ttx) wrote :
Monty Taylor (mordred)
Changed in openstack-ci:
milestone: none → grizzly
status: New → Confirmed
James E. Blair (corvus)
no longer affects: openstack-ci
Revision history for this message
Sean Dague (sdague) wrote :

This has been seen as recently as today:

Bug 1056213: Flakey build failures in some volume tests

First seen: 2013-01-29 15:59:12 UTC
Last seen: 2013-02-20 01:06:41 UTC
Rechecks: 8
Affecting projects:
openstack/glance
openstack/nova
openstack/keystone
Affecting changes:
20697
20699
18462
21216
21939
21564

Changed in tempest:
status: New → Confirmed
Changed in cinder:
status: New → Confirmed
Changed in cinder:
status: Confirmed → Invalid
Changed in tempest:
status: Confirmed → Invalid
Changed in cinder:
status: Invalid → Won't Fix
status: Won't Fix → Fix Committed
Changed in tempest:
status: Invalid → Fix Committed
Sean Dague (sdague)
summary: - Flakey build failures in some volume tests
+ Tempest failure because of LVM bug on precise
Revision history for this message
John Griffith (john-griffith) wrote :

So this bug was actually opened and was due to volume create and delete failures. There were two issue here, both have been addressed:

1. Volume delete on Precise would hang the system when attempting to do a secure delete
        FIX - Introduced "SECURE_DELETE=FALSE" as default in devstack and gate tests

2. Tempest was not waiting for delete when finishing volume tests, as a result there were many cases
     where the delete was slow and we'd attempt to create another volume but we were OUT of space on the Volume Group

        FIX - Added wait_for_resource_deletion to the tempest volume tests
        FIX - Increased the Volume Group size in devstack/tempest gate tests to 10G

Thierry Carrez (ttx)
Changed in cinder:
milestone: none → grizzly-3
status: Fix Committed → Fix Released
Sean Dague (sdague)
description: updated
Thierry Carrez (ttx)
Changed in cinder:
milestone: grizzly-3 → 2013.1
Sean Dague (sdague)
Changed in tempest:
milestone: none → havana-1
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.