github.com/kubernetes-incubator/kube-aws@v0.16.4/docs/advanced-topics/high-availability.md (about) 1 # High Availability 2 3 To achieve high availability using kube-aws, it is recommended to: 4 5 * Specify at least 3 for `etcd.count` in `cluster.yaml`. See [Optimal Cluster Size](https://coreos.com/etcd/docs/latest/v2/admin_guide.html#optimal-cluster-size) for details of etcd recommendations 6 * Specify at least 2 for `controller.count` in `cluster.yaml` 7 * Use 2 or more worker nodes, 8 * Avoid `t2.medium` or smaller instances for etcd and controller nodes. See [this issue](https://github.com/kubernetes-incubator/kube-aws/issues/138) for some additional discussion. 9 10 # Additional Reading 11 12 There's some additional documentation about [Building High-Availability Clusters](https://kubernetes.io/docs/admin/high-availability/) on the main Kubernetes documentation site. Although kube-aws will taken care of most of those concerns for you, it can be worth a read for a deeper understanding.