Created attachment 1014741 [details] Excerpt from evm.log Description of problem: When trying to terminate an instance, the instance gets terminated, but the task in CFME ends with an error and the instance remains in the database. Additional info: TripleO setup Version-Release number of selected component (if applicable): 5.4.0.0.19.20150410165622_ad23806 How reproducible: always Steps to Reproduce: 1. Have an instance in OpenStack 2. Terminate it Actual results: Instance gets terminated in the provider but then it shows an error in Configuration/Tasks: | Status = Error | 04/15/15 13:04:39 UTC | 04/15/15 13:04:33 UTC | Finished | unknown attribute: state | cirros2: 'vm_destroy' | admin | and the Instance is still shown in the UI (unless you do a manual refresh, then it gets archived) Expected results: Instance gets terminated and disappears
https://github.com/ManageIQ/manageiq/pull/2680
New commit detected on manageiq/master: https://github.com/ManageIQ/manageiq/commit/cc051cad27be10586636136bbd6cab8a88c6b6c6 commit cc051cad27be10586636136bbd6cab8a88c6b6c6 Author: Greg Blomquist <gblomqui> AuthorDate: Mon Apr 20 11:46:41 2015 -0400 Commit: Greg Blomquist <gblomqui> CommitDate: Mon Apr 20 11:46:41 2015 -0400 Destroying an OpenStack VM should not set state attr The state attribute on VMs is read only and used purely as a normalization layer in order to display a common state for all different types of VMs. When VMs are destroyed (or, otherwise acted upon), the appliance will set a temporary power state in order to reflect the impending change that's been submitted to the provider. The change submitted to the provider is handled largely asynchronously with any updates from the change coming in the form of events received after the fact. The temporary power state change when destroying OpenStack VMs was trying to set the state attribute. By changing that to the raw_power_state attribute, the power state is updated and reflected in the read only state attribute. https://bugzilla.redhat.com/show_bug.cgi?id=1212052 vmdb/app/models/vm_openstack/operations.rb | 2 +- vmdb/spec/models/vm_openstack_spec.rb | 17 +++++++++++++++++ 2 files changed, 18 insertions(+), 1 deletion(-)
Checked in 5.4.0.0.24.20150427192818_1fd9e49, now there is no error in the task message, the termination happens but there still is the suspended state which persists until next refresh. Is that intended? If yes, then I can move it to VERIFIED.
5.2.0.0.25 - instance gets archived correctly now. This was probably from the another BZ from power states but I wanted to be sure.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHBA-2015-1100.html