The following document describes the development of a cloud native cfs deployment on Kubernetes. Cloud native deployment indicates application understands that it is running within a cluster manager, and uses this cluster management infrastructure to help implement the application1.
You should have a Kubernetes cluster running and kubectl command line ready. Read getting started for upstream installation instructions.
You should know basics about Kubernetes concepts, including pods, service and replication controller. Basics tutorials explain them well.
Create a replication controller for cfs instances. Its default replica count is 1, so it creates single cfs instance in the cluster.
kubectl create -f cfs-controller.yaml
Check replication controller's status and pods' status:
kubectl describe rc cfs
kubectl describe pods --selector=name=cfs
Check cfs server in the pod is ready: [TODO: read and write from one cfs server]
kubectl logs cfs-xxxxx
Scale up cfs instances to 2:
kubectl scale rc cfs --replicas=2
[TODO: read and write from all cfs servers using cfs nameserver]
kubectl delete rc cfs