This epic tracks all the effort needed to deliver the solution related to the bug described below.

 Problem Description:

In RHOSP 17.1 ML2/OVN environment workloads lost network connectivity. Customer collected sosreport from problematic compute node and solved problem by restarting ovsdb-server.service.

In collected sosreport:

• there are no errors in /var/log/openvswitch/ovsdb-server.log while ps output tells that ovsdb-server process is running
• there are recurring errors in /var/log/containers/openvswitch/ovn-controller.log telling that connection to unix:/var/run/openvswitch/br-int.mgmt failed because of protocol error
• sosreport tool itself failed to collect various ovs outputs because of timeout (so it confirms that some problem is generally there)

Example error from /var/log/containers/openvswitch/ovn-controller.log:

2025-11-27T11:57:11.004Z|00354|rconn|WARN|unix:/var/run/openvswitch/br-int.mgmt: connection failed (Protocol error)

 Impact Assessment: problem is solved now, network connectivity was completely off when problem was there

 Software Versions: openvswitch3.3-3.3.6-141.el9fdp.x86_64

  Issue Type: bug

 Reproducibility: no reproducer

 Troubleshooting Actions: explained in description

 Logs: a single sosreport from compute node running affected ovsdb is attached to support case linked to this issue