This article will quickly guide you on resolution of the error “kube-system coredns crashloopbackoff” in kubernetes.

Basically coredns acts as as a basic DNS service in the kubernetes system. In case you want to explore more about this service please follow this link.

About the error:

Basically whenever you are firing up the command on the kubernetes server you are encountered with following “kube-system coredns crashloopbackoff” output.

$ kubectl get pods --all-namespaces
NAMESPACE     NAME                                 READY   STATUS             RESTARTS   AGE
kube-system   coredns-66bff467f8-mgqh2             0/1     CrashLoopBackOff   1          4m36s
kube-system   coredns-66bff467f8-wdtkr             0/1     CrashLoopBackOff   1          4m36s
kube-system   etcd-kubemaster                      1/1     Running            0          4m43s
kube-system   kube-apiserver-kubemaster            1/1     Running            0          4m43s
kube-system   kube-controller-manager-kubemaster   1/1     Running            0          4m43s
kube-system   kube-flannel-ds-amd64-7cn87          1/1     Running            0          39s
kube-system   kube-proxy-wgjk6                     1/1     Running            0          4m36s
kube-system   kube-scheduler-kubemaster            1/1     Running            0          4m43

Now in order to resolve this error simply follow below two steps:

Step1: Use the following command;

root@kubemaster$ kubectl -n kube-system edit configmap coredns

This will opened with open up with following output on the screeen as below:

# Please edit the object below. Lines beginning with a '#' will be ignored, 
# and an empty file will abort the edit. If an error occurs while saving this file will be 
# reopened with the relevant failures. 
# 
apiVersion: v1 
data: 
 Corefile: | 
   .:53 { 
         errors 
   health { 
       lameduck 5s 
    } 
    ready 
    kubernetes cluster.local in-addr.arpa ip6.arpa { 
       pods insecure 
    fallthrough in-addr.arpa ip6.arpa 
    ttl 30
   } prometheus :9153
   forward . /etc/resolv.conf 
   cache 30 
   #loop <------- 
   reload 
   loadbalance 
 } 
kind: ConfigMap

Now just find the “loop” word in it and hashed it or remove it. I just made arrow on the above output to make it clear for you. Save  and quit using normal vi command “:wq!”.

Step2: Now remove the old created coredns pods so that new one will be automatically created.

root@kubemaster$ kubectl -n kube-system delete pod -l k8s-app=kube-dns 
pod "coredns-66bff467f8-mgqh2" deleted 
pod "coredns-66bff467f8-wdtkr" deleted

That’s it you are done with the recovery simply wait for one or two minutes to recreate the coredns pods with running state as below;

root@kubemaster$ kubectl get pods --all-namespaces
NAMESPACE     NAME                                 READY   STATUS    RESTARTS   AGE
kube-system   coredns-66bff467f8-76wpx             1/1     Running   0          15s
kube-system   coredns-66bff467f8-rlz9j             1/1     Running   0          15s
kube-system   etcd-kubemaster                      1/1     Running   0          6m2s
kube-system   kube-apiserver-kubemaster            1/1     Running   0          6m2s
kube-system   kube-controller-manager-kubemaster   1/1     Running   0          6m2s
kube-system   kube-flannel-ds-amd64-7cn87          1/1     Running   0          118s
kube-system   kube-proxy-wgjk6                     1/1     Running   0          5m55s
kube-system   kube-scheduler-kubemaster            1/1     Running   0          6m2s