Bug 1470586

Summary: Instance Reconfigure does not work
Product: Red Hat CloudForms Management Engine Reporter: Aziza Karol <akarol>
Component: ProvidersAssignee: Scott Seago <sseago>
Status: CLOSED WORKSFORME QA Contact: Ola Pavlenko <opavlenk>
Severity: high Docs Contact:
Priority: high    
Version: 5.8.0CC: akarol, dberger, gblomqui, jfrey, jhardy, obarenbo, okolisny, opavlenk, tzumainn
Target Milestone: GA   
Target Release: cfme-future   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard: reconfigure:openstack
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2017-08-29 20:19:32 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: Openstack Target Upstream Version:
Embargoed:

Description Aziza Karol 2017-07-13 09:10:08 UTC
Description of problem:


Version-Release number of selected component (if applicable):
5.8.1.2.20170712143358_461cde4

How reproducible:
100%

Steps to Reproduce:
1.Add RHOS provider 'Compute-->Cloud-->Providers'
2.Navigate to 'Compute-->Cloud-->Instances'
3.Select and instance and select 'Configuration-->Reconfigure'
4.Change the Instance flavor and click submit

Actual results:
Only a flash message stating "Reconfiguring Instance "xxx" from m1.tiny to m1.medium is displayed but it does not get changed.

Also there is no service request created for this reconfiguration request in service->request page.

Expected results:
reconfiguration of instance from one flavor to another should be successful.

Additional info:
No log details in the automation.log as well.

automation.log
----] I, [2017-07-13T05:00:37.775192 #5712:5cd140]  INFO -- : Instantiating [/System/Process/Event?EventStream%3A%3Aevent_stream=418&ExtManagementSystem%3A%3Aext_management_system=3&MiqEvent%3A%3Amiq_event=418&MiqServer%3A%3Amiq_server=2&User%3A%3Auser=1&event_stream_id=418&event_type=ems_auth_valid&miq_event_id=418&object_name=Event&vmdb_object_type=ext_management_system]
[----] I, [2017-07-13T05:00:37.798139 #5712:5cd140]  INFO -- : Updated namespace [/System/Process/Event?EventStream%3A%3Aevent_stream=418&ExtManagementSystem%3A%3Aext_management_system=3&MiqEvent%3A%3Amiq_event=418&MiqServer%3A%3Amiq_server=2&User%3A%3Auser=1&event_stream_id=418&event_type=ems_auth_valid&miq_event_id=418&object_name=Event&vmdb_object_type=ext_management_system  ManageIQ/System]
[----] I, [2017-07-13T05:00:37.843488 #5712:5cd140]  INFO -- : Following Relationship [miqaedb:/System/Event/MiqEvent/POLICY/ems_auth_valid#create]
[----] I, [2017-07-13T05:00:37.862130 #5712:5cd140]  INFO -- : Updated namespace [miqaedb:/System/Event/MiqEvent/POLICY/ems_auth_valid#create  ManageIQ/System/Event/MiqEvent]
[----] I, [2017-07-13T05:00:37.871143 #5712:5cd140]  INFO -- : Instance [/ManageIQ/System/Event/MiqEvent/POLICY/ems_auth_valid] not found in MiqAeDatastore - trying [.missing]
[----] I, [2017-07-13T05:00:37.881073 #5712:5cd140]  INFO -- : Following Relationship [miqaedb:/System/event_handlers/event_enforce_policy#create]
[----] I, [2017-07-13T05:00:37.884477 #5712:5cd140]  INFO -- : Updated namespace [miqaedb:/System/event_handlers/event_enforce_policy#create  ManageIQ/System]
[----] I, [2017-07-13T05:00:37.892001 #5712:5cd140]  INFO -- : Updated namespace [System/event_handlers/event_enforce_policy  ManageIQ/System]
[----] I, [2017-07-13T05:00:37.896591 #5712:5cd140]  INFO -- : Invoking [builtin] method [/ManageIQ/System/event_handlers/event_enforce_policy] with inputs [{}]
[----] I, [2017-07-13T05:00:37.907219 #5712:5cd140]  INFO -- : Followed  Relationship [miqaedb:/System/event_handlers/event_enforce_policy#create]
[----] I, [2017-07-13T05:00:37.907603 #5712:5cd140]  INFO -- : Followed  Relationship [miqaedb:/System/Event/MiqEvent/POLICY/ems_auth_valid#create]
[----] I, [2017-07-13T05:00:59.889142 #5702:5cd140]  INFO -- : MiqAeEvent.build_evm_event >> event=<"evm_worker_start"> inputs=<{:event_details=>"Worker started: ID [216], PID [35307], GUID [c971bcd6-67a9-11e7-98a2-001a4a4809f1]", :type=>"ManageIQ::Providers::StorageManager::CinderManager::EventCatcher", "MiqEvent::miq_event"=>419, :miq_event_id=>419, "EventStream::event_stream"=>419, :event_stream_id=>419}>
[----] I, [2017-07-13T05:01:05.097628 #5712:5cd140]  INFO -- : Instantiating [/System/Process/Event?EventStream%3A%3Aevent_stream=419&MiqEvent%3A%3Amiq_event=419&MiqServer%3A%3Amiq_server=2&User%3A%3Auser=1&event_details=Worker%20started%3A%20ID%20%5B216%5D%2C%20PID%20%5B35307%5D%2C%20GUID%20%5Bc971bcd6-67a9-11e7-98a2-001a4a4809f1%5D&event_stream_id=419&event_type=evm_worker_start&miq_event_id=419&object_name=Event&type=ManageIQ%3A%3AProviders%3A%3AStorageManager%3A%3ACinderManager%3A%3AEventCatcher&vmdb_object_type=miq_server]
[----] I, [2017-07-13T05:01:05.123458 #5712:5cd140]  INFO -- : Updated namespace [/System/Process/Event?EventStream%3A%3Aevent_str

Comment 3 Tzu-Mainn Chen 2017-07-13 20:39:57 UTC
Scott, can you verify this while looking into adding the approval process into reconfigure? Thanks!

Comment 4 Scott Seago 2017-08-11 15:03:05 UTC
I'm starting to suspect that this may not be a bug at all.

To address the last part first -- there is no service request created because the action is called directly right now. There's an existing development task to modify this to use the request/approval process, but that is not supported in the current release.

Regarding the failed reconfiguration, can you see if reconfiguration works on the underlying OpenStack instance -- run directly via CLI?

Reconfiguration was failing for me in the underlying OpenStack CLI without triggering an OpenStack error -- the CLI said "success", so that's a possible OpenStack bug rather than a CloudForms bug -- if OpenStack says it succeeded, then CloudForms will report the same.

What I discovered was that my compute nodes didn't have the SSH keys set up to allow passwordless SSH to succeed between the nodes, which prevented reconfiguration from working (again this is an OpenStack configuration issue rather than a CLoudForms issue).

Once reconfiguration works on the OpenStack side, try it again via CloudForms and see if it works properly there.

Comment 5 Aziza Karol 2017-08-18 04:35:22 UTC
This issues seems to be specific to rhos provider as for other providers the reconfiguration works well.

We should have Openstack QE check this. Ola can you investigate on this issue?

Comment 6 Tzu-Mainn Chen 2017-08-29 14:12:29 UTC
I think we need someone from QE to verify Scott's theory - Ola, would that be you or someone else... ?

Comment 7 Oleksandr Kolisnyk 2017-08-29 20:02:21 UTC
I've checked on 5.8.2.0 with RHOS 11 - reconfiguration is being finished successfully both from OSP and CFME sides.

Comment 8 Tzu-Mainn Chen 2017-08-29 20:19:32 UTC
Thanks! I'm closing as WORKSFORME then - feel free to re-open if the problem persists, but be aware that the issue may be on the OpenStack side.