Bug 1947385

Summary: [ovn] ovsdbapp doesn't seem to ever reconnect to OVN databases sometimes
Product: Red Hat OpenStack Reporter: Daniel Alvarez Sanchez <dalvarez>
Component: python-ovsdbappAssignee: Terry Wilson <twilson>
Status: CLOSED CURRENTRELEASE QA Contact: Toni Freger <tfreger>
Severity: high Docs Contact:
Priority: unspecified    
Version: 16.1 (Train)CC: ekuris, eolivare, jlibosva, jmelvin, oblaut, twilson
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: python-ovsdbapp-0.17.4-1.20201114022230.el8ost Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2021-04-12 07:41:32 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:

Description Daniel Alvarez Sanchez 2021-04-08 11:09:09 UTC 
We faced a problem when, upon a failover of the OVN DBs (Active/Passive) in OpenStack, some neutron-server workers seemed to think that they were connected to the OVN DBs but they were actually not connected.

From that point on, the logs in neutron-server show that many workers will never be able to execute a transaction and for some reason, either ovsdbapp or the reconnection FSM in the Python OVS library won't detect this condition and they won't ever attempt to reconnect.

ss on the PID shows that the connection is not established.

The only solution is to kill the worker(s) or restart neutron-server.
Comment 1 Jakub Libosvar 2021-04-12 07:41:32 UTC 
The fix is available in OSP 16.1.5