3.1/3.2 standup cluster issue: NotReady: cni plugin not initialized

mzodun

Following the guide several times over, I am unable to stand up my cluster. I have VMWare 8 with two VMS each 2vcpu, 8GB RAM, 100GB HDD running Ubuntu 20.04.6LTS.

I ran the kubeadm init script, as well as the kubectl apply -f cilium-cni.yaml

My VM network is 192.168.4.x/24. I change the podSubnet in kubeadm-config.yaml to 10.200.0.0/16 and matched that in cilium-cni.yaml. I have tried making them all the same, and making them different no luck.

NAME STATUS ROLES AGE VERSION
cp NotReady control-plane 3m34s v1.31.1
k8sworker.st.lab NotReady 2m14s v1.31.1

kubectl describe node shows the following:
Ready False Mon, 14 Apr 2025 13:51:39 +0000 Mon, 14 Apr 2025 13:51:29 +0000 KubeletNotReady container runtime network not ready: NetworkReady=false reason:NetworkPluginNotReady message:Network plugin returns error: cni plugin not initialized

kubectcl get pods -A shows this:
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-system cilium-envoy-d8gc5 1/1 Running 0 4m51s
kube-system cilium-envoy-kt5xm 1/1 Running 0 5m20s
kube-system cilium-f5lz9 0/1 Init:0/6 3 (76s ago) 4m51s
kube-system cilium-operator-5c7867ccd5-hnz4b 0/1 Running 3 (46s ago) 5m20s
kube-system cilium-operator-5c7867ccd5-v458p 0/1 Running 3 (78s ago) 5m20s
kube-system cilium-rk2qk 0/1 Init:CrashLoopBackOff 3 (16s ago) 5m20s
kube-system coredns-7c65d6cfc9-f8gpx 0/1 Pending 0 6m3s
kube-system coredns-7c65d6cfc9-hv4ws 0/1 Pending 0 6m3s
kube-system etcd-cp 1/1 Running 4 6m8s
kube-system kube-apiserver-cp 1/1 Running 4 6m8s
kube-system kube-controller-manager-cp 1/1 Running 0 6m8s
kube-system kube-proxy-xnqsb 1/1 Running 0 6m3s
kube-system kube-proxy-zz7t7 1/1 Running 0 4m51s
kube-system kube-scheduler-cp 1/1 Running 4 6m8s

Any help is appreciated.

chrispokorni

Hi @mzodun,

Your VMs seem so to be adequately sized, however, the OS version seems to be out of sync with the most recent release of the lab guide.
How is the 100 GB HDD disk space allocated per VM - dynamic or fully pre-allocated?
How many and what type of network interfaces do you have per VM (nat, host, bridge, etc...)?
Also, have you enabled all ingress traffic to each VM? That is all protocols from all sources to all port destinations?

Regards,
-Chris

mzodun

I was able to stand up my cluster after reinstall with Ubuntu 24.04 (when it did not initially work with 24.04 I switched to 20.04). I think the actual issue was enabling promiscuous mode on the vSwitch in VMWare. Thank you.