coredns gets killed by oom

[chrifey]

I deployed a cluster with 3 nodes (master + 2 worker nodes).

After the deployment is finished, the cluster is fine, but when I schedule services, the two coredns pods get killed from out-of-memory (oom). That makes coredns going into a crash-loop.

I'm not sure, whether this I a config problem on my side, but I fixed it with using kube-dns [1] instead of coredns by passing --feature-gates=CoreDNS=false.

[1] https://kubernetes.io/docs/tasks/administer-cluster/coredns/#installing-kube-dns-instead-of-coredns-with-kubeadm

Docker Version: 18.04.0-ce K8s: v1.11.4

[chris-short]

Is this still an issue?