I have setup a incus ovn network with 3-node cluster each running ovn-controller and ovn setup as defined in the guide.
I face a regular issue that container IP addresses are gone and I need to restart the ovn-controller, ovn-central on each machine to make it work again. This shows an unstable ovn network. I tried to troubleshoot it but there are no network traces or logs where I can find the error. Also incus do not record any logs of it. All the containers continue to have IPv6 but IPv4 addresses are gone. It seems something wrong but cannot pinpoint.
Issue is that intermittent and unstable ovn leads to unreliable container access which for users is an issue.
Is there a way to make it better some guide on it?