kilo: k3s kilo pods crashlooping

sudo kubectl apply -f https://raw.githubusercontent.com/squat/kilo/master/manifests/kilo-k3s-flannel.yaml

serviceaccount/kilo created
clusterrole.rbac.authorization.k8s.io/kilo created
clusterrolebinding.rbac.authorization.k8s.io/kilo created
daemonset.apps/kilo created

sudo kubectl logs -f kilo-cz64w -n kube-system

failed to create Kubernetes config: Error loading config file "/etc/kubernetes/kubeconfig": read /etc/kubernetes/kubeconfig: is a directory

I think the problem is with kilo-k3s-flannel.yaml:99.

About this issue

Original URL
State: closed
Created 4 years ago
Comments: 24 (16 by maintainers)

Most upvoted comments

Hi @squat, I got it working after re-doing all the steps in #11 with some modifications. But, I ended up using kilo-k3s.yaml because kilo-k3s-flannel.yaml did not work for some reason. I will try again today. In either files, I had to update the kubeconfig volume path:

❯ git diff manifests/kilo-k3s.yaml
diff --git a/manifests/kilo-k3s.yaml b/manifests/kilo-k3s.yaml
index 34c0572..6b8b5aa 100644
--- a/manifests/kilo-k3s.yaml
+++ b/manifests/kilo-k3s.yaml
@@ -158,4 +158,4 @@ spec:
           path: /var/lib/kilo
       - name: kubeconfig
         hostPath: 
-          path: /var/lib/rancher/k3s/agent/kubeconfig.yaml
+          path: /etc/rancher/k3s/k3s.yaml

I also had to copy /etc/rancher/k3s/k3s.yaml to the other nodes, because I dont think they are copied anymore as you say in #11.

Using k3s v1.0.1

ibalajiarun on Jan 4, 2020