deploymentconfig

Documentation

Index

• type DeploymentConfigController
  • func NewDeploymentConfigController(dcInformer, rcInformer, podInformer cache.SharedIndexInformer, ...) *DeploymentConfigController
  • func (c *DeploymentConfigController) Handle(config *deployapi.DeploymentConfig) error
  • func (c *DeploymentConfigController) Run(workers int, stopCh <-chan struct{})

Types

type DeploymentConfigController

type DeploymentConfigController struct {
	// contains filtered or unexported fields
}

DeploymentConfigController is responsible for creating a new deployment when:

1. The config version is > 0 and,
2. No deployment for the version exists.

The controller reconciles deployments with the replica count specified on the config. The active deployment (that is, the latest successful deployment) will always be scaled to the config replica count. All other deployments will be scaled to zero.

If a new version is observed for which no deployment exists, any running deployments will be cancelled. The controller will not attempt to scale running deployments.

func NewDeploymentConfigController

func NewDeploymentConfigController(dcInformer, rcInformer, podInformer cache.SharedIndexInformer, oc osclient.Interface, kc kclientset.Interface, codec runtime.Codec) *DeploymentConfigController

NewDeploymentConfigController creates a new DeploymentConfigController.

func (*DeploymentConfigController) Handle

func (c *DeploymentConfigController) Handle(config *deployapi.DeploymentConfig) error

Handle implements the loop that processes deployment configs. Since this controller started using caches, the provided config MUST be deep-copied beforehand (see work() in factory.go).

func (*DeploymentConfigController) Run

func (c *DeploymentConfigController) Run(workers int, stopCh <-chan struct{})

Run begins watching and syncing.