kubernetes example Kenmo
k8s is an open source, cloud management platform for application of the container on multiple hosts, Kubernetes goal is to make application deployment container of simple and effective (powerful), Kubernetes provides application deployment, planning, updates, maintenance a mechanism. The project relies on engineering practice based on adaptive k8s do rewrite domestic chip.
Blog only present on a large frame of k8s analyzed by modeling cases, on which the security, load balancing analysis details not.
First, the general framework
Most concepts such k8s Node, Pod, service and so can be seen as a resource objects, kubectl tools almost all resource objects are provided by performing k8s add, delete, change, and other operations stored in the etcd persistent storage.
, Can be analyzed from a business model point of view mainly for master master node, node node and pod
Two, master
master refers to the cluster control node, is responsible for the management and control of the entire cluster, basically k8s all control commands are sent to it. We are basically behind the command to run on the master node. Usually it will occupy a stand-alone x86 servers (or a virtual machine).
master to achieve the main business functions:
k8s API server (kube-apiserver): provides a key service process HTTP Rest interface, is the only entrance entrance process additions and deletions to change search all resources, as well as a cluster of cluster control. kubectl command calls to api server, check the resources additions and deletions to achieve change.
kube-controller-manager: Automation Control Center k8s all resource objects.
kube-scheduler: pod scheduling process.
etcd server process: the storage object persistence and data synchronization
Three, node
k8s machines other cluster node is called a node, the Node may be a single physical machine, or may be a virtual machine. When a node is down, the workload will be automatically transferred on the master to other nodes.
node node main achievement of business functions:
kubelet: responsible pod corresponding container to create, start and stop other tasks.
kube-porxy: for essential component communication service.
docker engine: docker engine, the unit responsible for the creation and management of container.
node node in the cluster can be dynamically added to k8s during operation, by default kubelet master will register itself, and regularly report their information to the master.
Fourth, the complete use case diagram
