dosubot[bot]
commented
5 days ago Hey @kvaps! I'm here to help you with any bugs, questions, or contributions you have. Let's tackle this issue together!
The issue you're experiencing with gonetworkmanager.NewNetworkManager() getting stuck during boot on Talos Linux could be related to dependencies such as the DBus connection or the NetworkManager service. The function relies on a DBus connection to communicate with the NetworkManager service, and if there are issues with the DBus service or if the NetworkManager service is not running, it could cause the function to get stuck [1].
Additionally, the readiness probe for kube-ovn-cni pods checks the operational state of the ovn0 network interface and network connectivity to the gateway. If these checks fail, the pods will not reach a ready state [2].
To address this, ensure that the NetworkManager service is running and that there are no issues with the DBus connection on your Talos Linux setup. You might also want to verify the configuration specific to Talos Linux, such as declaring the openvswitch module and using the correct Helm chart options [3].
To continue talking to Dosu, mention @dosu.
zhangzujian
kvaps
Kube-OVN Version
v1.12.19
Kubernetes Version
v1.30.3
Operation-system/Kernel Version
Description
Hey I faced that some issue that kube-ovn get stuck on the boot. I spent a few hours for debugging and just wanted to report an issue.
I found out that it stucks on calling this function: https://github.com/kubeovn/kube-ovn/blob/c414e6b967a3b985365f2e546b2afbd51d258f0c/pkg/daemon/nm_linux.go#L34
Steps To Reproduce
Someone may face with this on Talos Linux
Current Behavior
kube-ovn-cni pods do not become to ready state, due to readiness probe, logs say:
Expected Behavior
I expect message:
and kube-ovn-cni become to ready