Description of problem: Upon a fencing or crash of a controller node, the node reboots but the bonds (bond0 and bond1) never come up by themselves. We need to restart the network service to make things work again. If client does a normal reboot, no problem is seen. Client setup is with OVN. And bonds are OVS bond and not Linux bond. Is the source of the issue related with using OVS bond instead of Linux bond ? I was told to run the following commands next time they can recreate the issue for troubleshooting: ovs-ofctl dump-flows/br-ex ovs-ofctl dump-flows/br-storage I have logs from when it happened. Will share details in next private comment. Version-Release number of selected component (if applicable): OSP13 z12 ovn2.11-2.11.1-33.el7fdp.x86_64 How reproducible: Often if not 100% Steps to Reproduce: 1. Fence/Crash the OS. 2. 3. Actual results: Network doesnt work, need to restart network service to make things work. Expected results: Work without restarting anything. Additional info: I have /var/log/ from the controller when it was fenced.
Closing this BZ since it is a duplicate of https://bugzilla.redhat.com/show_bug.cgi?id=1957025 Thank you for the help. *** This bug has been marked as a duplicate of bug 1957025 ***