Is it possible to enable autoscaling of nodes (minions) in Kubernetes running on CoreOS in OpenStack?
I only read about AWS and GCE.
You'd probably need to deploy Kubernetes with Heat and use autoscale in your template. Example:
http://superuser.openstack.org/articles/simple-auto-scaling-environment-with-heat
and a template for Kubernetes to build on:
https://github.com/metral/corekube/blob/master/corekube-openstack.yaml
there are two ways to autoscale k8s nodes in cloud provider(such as openstack, AWS), first is cloud provider autoscale cluster nodes according to k8s resource utilizing, in this situation, cloud provider know k8s is running in cloud, cloud provider is initiator, the related project is cluster-api and cluster autoscale. Then the second way is k8s ask VM from cloud provider and then the VMs join k8s, in this situation, k8s is initiator and cloud provider doesn't know k8s is running on it.