Bug 1076100

Summary:	nova scrub doesn't remove a vm
Product:	Red Hat OpenStack	Reporter:	Jaroslav Henner <jhenner>
Component:	openstack-nova	Assignee:	Ryan Hallisey <rhallise>
Status:	CLOSED NOTABUG	QA Contact:	Ami Jeain <ajeain>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	3.0	CC:	jhenner, ndipanov, sgordon, yeylon
Target Milestone:	ga
Target Release:	6.0 (Juno)
Hardware:	x86_64
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2014-10-16 15:07:15 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:	1083620
Bug Blocks:

Description Jaroslav Henner 2014-03-13 14:34:29 UTC

Description of problem:
Checking the nova code, nova scrub should disassociate the network associated with the given tenant, but it fails to do so. There is no error in log.


Version-Release number of selected component (if applicable):
openstack-nova-api.noarch 2013.2.1-1.el6ost @puddle
openstack-nova-cert.noarch 2013.2.1-1.el6ost @puddle
openstack-nova-common.noarch 2013.2.1-1.el6ost @puddle
openstack-nova-compute.noarch 2013.2.1-1.el6ost @puddle
openstack-nova-conductor.noarch 2013.2.1-1.el6ost @puddle
openstack-nova-console.noarch 2013.2.1-1.el6ost @puddle
openstack-nova-network.noarch 2013.2.1-1.el6ost @puddle
openstack-nova-novncproxy.noarch 2013.2.1-1.el6ost @puddle
openstack-nova-objectstore.noarch 2013.2.1-1.el6ost @puddle
openstack-nova-scheduler.noarch 2013.2.1-1.el6ost @puddle


How reproducible:
100%


Steps to Reproduce:
Actual results:

nova-manage --nodebug network list | head
keystone tenant-create --name foo
keystone user-role-add --user admin --role Member --tenant foo
OS_TENANT_NAME=foo nova boot --image cirros-0.3.1-x86_64-disk.vmdk --flavor m1.tiny foo_vm

nova list --all-tenants
nova-manage network list | head

nova scrub foo
nova list --all-tenants
nova-manage --nodebug network list | head

######## The network is still associated in this point.

keystone tenant-delete foo
nova-manage --nodebug network list | head
nova list --all-tenants

######## Here as well

Comment 2 Stephen Gordon 2014-03-26 17:14:05 UTC

Hi Jaroslav,

Unclear if it matters in this case but was this when you were testing with the vCenter driver, or KVM?

Thanks,

Steve

Comment 3 Jaroslav Henner 2014-03-27 08:27:44 UTC

(In reply to Stephen Gordon from comment #2)
> Hi Jaroslav,
> 
> Unclear if it matters in this case but was this when you were testing with
> the vCenter driver, or KVM?
> 
> Thanks,
> 
> Steve

I think this was with the vCenter (obviously there was vmdk image used), but i also think that it doesn't matter as IIRC the code doing the nova scrub is buggy general. I can retest it with libvirt with no, problem.

Comment 4 Ryan Hallisey 2014-03-28 17:10:00 UTC

Can you reproduce this in libvirt?  Also, are you using nova network?

Comment 5 Jaroslav Henner 2014-04-02 15:16:10 UTC

I have reproduced it with libvirt and made 1083620. Resolution of this bug depends on what we decide to do with 1083620.

Comment 6 Jaroslav Henner 2014-09-10 15:48:05 UTC

I believe all further discussion should be done on the LP bug.

Comment 8 Russell Bryant 2014-10-16 15:07:15 UTC

"nova scrub" is a very limited command.  All it tries to do is delete networks and security groups.  It doesn't try to delete instances (by design).

I updated the command help text to try to make this more clear:

https://review.openstack.org/#/c/126975/