[OVS] When cleaning min QoS, queue could be already deleted
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
neutron |
Fix Released
|
High
|
Rodolfo Alonso |
Bug Description
When OVS agent is started and QoS extension is enabled, the minimum bandwidth QoS and Queues are deleted [1]-->[2].
Although is not possible to have a QoS with deleted Queues assigned (a "referential integrity violation" will be risen), a race condition could happen during functional and fullstack tests [3].
While deleting the Queues referred in the Qos register, an exception catch should be added to hide if the Queue is already deleted. In this case there is no need to raise this exception.
[1] https:/
[2] https:/
[3] http://
Changed in neutron: | |
assignee: | nobody → Rodolfo Alonso (rodolfo-alonso-hernandez) |
tags: | added: ovs qos |
Changed in neutron: | |
importance: | Undecided → High |
tags: | added: neutron-proactive-backport-potential |
tags: | removed: neutron-proactive-backport-potential |
Fix proposed to branch: master /review. openstack. org/642501
Review: https:/