I have a 3 node kube control network setup with etcd
, kube-apiserver
, kube-control-manager
and kube-scheduler
running. When I check the status of the components it says etcd is unhealthy, but etcd reports that is healthy
> kubectl get componentstatus --kubeconfig .kube/admin.kubeconfig
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME STATUS MESSAGE ERROR
etcd-0 Unhealthy HTTP probe failed with statuscode: 404
etcd-2 Unhealthy HTTP probe failed with statuscode: 404
etcd-1 Unhealthy HTTP probe failed with statuscode: 404
scheduler Healthy ok
controller-manager Healthy ok
And when I check the status of etcd I get the below result
> etcdctl endpoint health --write-out=table
+------------------------+--------+------------+-------+
| ENDPOINT | HEALTH | TOOK | ERROR |
+------------------------+--------+------------+-------+
| http://10.0.1.164:2379 | true | 2.867446ms | |
| http://10.0.1.62:2379 | true | 3.305321ms | |
| http://10.0.1.176:2379 | true | 6.970254ms | |
+------------------------+--------+------------+-------+
I am also able to put and retrieve items from etcd, the endpoints provided to kube-apiserver are also correct. How do I debug an issue like this?