I'm running kubernetes with the supposed pod network cidr of Everything seems to agree with that:

# history | grep kubeadm
kubeadm init --control-plane-endpoint "guide:6443" --upload-certs --pod-network-cidr=

$ kubectl get nodes -o kubectl get nodes -o jsonpath='{.items[*].spec.podCIDR}'

# cat /run/flannel/subnet.env 

However all the pods I create get IPs in the range.

$ kubectl get pods -o wide
NAME                                   READY   STATUS             RESTARTS   AGE   IP            NODE         NOMINATED NODE   READINESS GATES
busybox1                               1/1     Running            1          13m   k8s-node-1   <none>           <none>
debain-shell-547f8fc54b-v8p99          0/1     ImagePullBackOff   0          27m    k8s-node-1   <none>           <none>
nginx-deployment-ex-54f57cf6bf-k55nn   1/1     Running            1          27m    k8s-node-1   <none>           <none>
nginx-deployment-ex-54f57cf6bf-q2khx   1/1     Running            1          27m    k8s-node-1   <none>           <none>
nginx-deployment-ex-54f57cf6bf-r8b7v   1/1     Running            1          27m    k8s-node-1   <none>           <none>

I can't talk to coredns (which is how I found this condition) and I can't seem to use my ingress service anymore (though I've restarted enough things by now I may have just temporarily broken that).

What is causing this, or alternatively, where should I look for more info?


This was caused by missing --network-plugin=cni on the end of kubelet. Added that to the args and it all works now.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.