masterserver

README

Main node service function description

• Working model

• The master node can have multiple working points. In principle, the instance of the cloud help management node works as the master node model.

• The work task of the master node is based on ETCD to achieve uniqueness. For example, processing the execution record of a task is completed by only one Master instance.

• Job creation

• When the Task starts to schedule execution, a Job object is created and executed based on the ETCD notification to the execution node.

• Process job execution results

• Installation task result processing, serial group installation tasks, global configuration, etc.

• Detect task result processing, trigger the next task execution according to the strategy, or user alarm.

• Handle group task execution.

Documentation

Index

• type MasterServer
  • func NewMasterServer(modelnode *client.HostNode, kubecli kubecache.KubeClient) (*MasterServer, error)
  • func (m *MasterServer) GetRegistry() *prometheus.Registry
  • func (m *MasterServer) Start(errchan chan error) error
  • func (m *MasterServer) Stop(i interface{})

Types

type MasterServer

type MasterServer struct {
	*store.Client
	*client.HostNode
	Cluster *node.Cluster
	// contains filtered or unexported fields
}

MasterServer

func NewMasterServer

func NewMasterServer(modelnode *client.HostNode, kubecli kubecache.KubeClient) (*MasterServer, error)

NewMasterServer

func (*MasterServer) GetRegistry

func (m *MasterServer) GetRegistry() *prometheus.Registry

GetRegistry

func (*MasterServer) Start

func (m *MasterServer) Start(errchan chan error) error

Start master node

func (*MasterServer) Stop

func (m *MasterServer) Stop(i interface{})

Stop