Bug 891408

Summary: [RFE] If the compute node running a VM dies, the state in the database remains Active:Running indefinitely
Product: Red Hat OpenStack Reporter: Perry Myers <pmyers>
Component: openstack-novaAssignee: Nikola Dipanov <ndipanov>
Status: CLOSED DUPLICATE QA Contact: yeylon <yeylon>
Severity: medium Docs Contact:
Priority: medium    
Version: 2.0 (Folsom)CC: eglynn, mmagr, sgordon, srevivo
Target Milestone: ---Keywords: FutureFeature, Improvement, MoveUpstream, Triaged
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
See Also: https://launchpad.net/bugs/1072734
Fixed In Version: Doc Type: Enhancement
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2016-01-28 15:49:15 EST Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Description Perry Myers 2013-01-02 15:51:14 EST
Description of problem:
If you run a VM on a compute node, and that compute node dies (hardware failure or just a simple shutdown) the state of the running VM is never updated in the database.  It'll remain as Active:Running indefinitely and it is left to the administrator of the cloud to clean up the state manually.  From a user PoV this is not ideal.

If the compute node comes back online, then it should update the state periodically to reconcile the actual state with the db state.  But if that compute node is dead for a long time, this state could remain out of sync for a while.

Need some sort of process that knows when to set the status for a VM to be stale/unknown.
Comment 1 Martin Magr 2013-01-15 10:05:35 EST
Setting as RFE as it seems that this issue is known to upstream [1]. So we should either wait till somebody will implement it or implement it ourselves.

[1] - https://bugs.launchpad.net/nova/+bug/1072734
Comment 6 Stephen Gordon 2016-01-28 15:49:15 EST

*** This bug has been marked as a duplicate of bug 1178665 ***