apiclient

package
v1.14.10-beta.0 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Nov 13, 2019 License: Apache-2.0 Imports: 33 Imported by: 0

Documentation

Index

Constants

This section is empty.

Variables

This section is empty.

Functions

func CreateOrRetainConfigMap added in v1.11.0 

func CreateOrRetainConfigMap(client clientset.Interface, cm *v1.ConfigMap, configMapName string) error

CreateOrRetainConfigMap creates a ConfigMap if the target resource doesn't exist. If the resource exists already, this function will retain the resource instead.

func CreateOrUpdateClusterRole 

func CreateOrUpdateClusterRole(client clientset.Interface, clusterRole *rbac.ClusterRole) error

CreateOrUpdateClusterRole creates a ClusterRole if the target resource doesn't exist. If the resource exists already, this function will update the resource instead.

func CreateOrUpdateClusterRoleBinding 

func CreateOrUpdateClusterRoleBinding(client clientset.Interface, clusterRoleBinding *rbac.ClusterRoleBinding) error

CreateOrUpdateClusterRoleBinding creates a ClusterRoleBinding if the target resource doesn't exist. If the resource exists already, this function will update the resource instead.

func CreateOrUpdateConfigMap 

func CreateOrUpdateConfigMap(client clientset.Interface, cm *v1.ConfigMap) error

CreateOrUpdateConfigMap creates a ConfigMap if the target resource doesn't exist. If the resource exists already, this function will update the resource instead.

func CreateOrUpdateDaemonSet 

func CreateOrUpdateDaemonSet(client clientset.Interface, ds *apps.DaemonSet) error

CreateOrUpdateDaemonSet creates a DaemonSet if the target resource doesn't exist. If the resource exists already, this function will update the resource instead.

func CreateOrUpdateDeployment 

func CreateOrUpdateDeployment(client clientset.Interface, deploy *apps.Deployment) error

CreateOrUpdateDeployment creates a Deployment if the target resource doesn't exist. If the resource exists already, this function will update the resource instead.

func CreateOrUpdateRole 

func CreateOrUpdateRole(client clientset.Interface, role *rbac.Role) error

CreateOrUpdateRole creates a Role if the target resource doesn't exist. If the resource exists already, this function will update the resource instead.

func CreateOrUpdateRoleBinding 

func CreateOrUpdateRoleBinding(client clientset.Interface, roleBinding *rbac.RoleBinding) error

CreateOrUpdateRoleBinding creates a RoleBinding if the target resource doesn't exist. If the resource exists already, this function will update the resource instead.

func CreateOrUpdateSecret 

func CreateOrUpdateSecret(client clientset.Interface, secret *v1.Secret) error

CreateOrUpdateSecret creates a Secret if the target resource doesn't exist. If the resource exists already, this function will update the resource instead.

func CreateOrUpdateServiceAccount 

func CreateOrUpdateServiceAccount(client clientset.Interface, sa *v1.ServiceAccount) error

CreateOrUpdateServiceAccount creates a ServiceAccount if the target resource doesn't exist. If the resource exists already, this function will update the resource instead.

func DefaultMarshalFunc 

func DefaultMarshalFunc(obj runtime.Object, gv schema.GroupVersion) ([]byte, error)

DefaultMarshalFunc is the default MarshalFunc used; uses YAML to print objects to the user

func DeleteDaemonSetForeground 

func DeleteDaemonSetForeground(client clientset.Interface, namespace, name string) error

DeleteDaemonSetForeground deletes the specified DaemonSet in foreground mode; i.e. it blocks until/makes sure all the managed Pods are deleted

func DeleteDeploymentForeground added in v1.9.0 

func DeleteDeploymentForeground(client clientset.Interface, namespace, name string) error

DeleteDeploymentForeground deletes the specified Deployment in foreground mode; i.e. it blocks until/makes sure all the managed Pods are deleted

func NewDryRunClient 

func NewDryRunClient(drg DryRunGetter, w io.Writer) clientset.Interface

NewDryRunClient is a wrapper for NewDryRunClientWithOpts using some default values

func NewDryRunClientWithOpts 

func NewDryRunClientWithOpts(opts DryRunClientOptions) clientset.Interface

NewDryRunClientWithOpts returns a clientset.Interface that can be used normally for talking to the Kubernetes API. This client doesn't apply changes to the backend. The client gets GET/LIST values from the DryRunGetter implementation. This client logs all I/O to the writer w in YAML format

func PatchNode added in v1.11.0 

func PatchNode(client clientset.Interface, nodeName string, patchFn func(*v1.Node)) error

PatchNode tries to patch a node using patchFn for the actual mutating logic. Retries are provided by the wait package.

func PatchNodeOnce added in v1.13.0 

func PatchNodeOnce(client clientset.Interface, nodeName string, patchFn func(*v1.Node)) func() (bool, error)

PatchNodeOnce executes patchFn on the node object found by the node name. This is a condition function meant to be used with wait.Poll. false, nil implies it is safe to try again, an error indicates no more tries should be made and true indicates success.

func PrintBytesWithLinePrefix 

func PrintBytesWithLinePrefix(w io.Writer, objBytes []byte, linePrefix string)

PrintBytesWithLinePrefix prints objBytes to writer w with linePrefix in the beginning of every line

func TryRunCommand 

func TryRunCommand(f func() error, failureThreshold int) error

TryRunCommand runs a function a maximum of failureThreshold times, and retries on error. If failureThreshold is hit; the last error is returned

Types

type ClientBackedDryRunGetter 

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

ClientBackedDryRunGetter implements the DryRunGetter interface for use in NewDryRunClient() and proxies all GET and LIST requests to the backing API server reachable via rest.Config

func NewClientBackedDryRunGetter 

func NewClientBackedDryRunGetter(config *rest.Config) (*ClientBackedDryRunGetter, error)

NewClientBackedDryRunGetter creates a new ClientBackedDryRunGetter instance based on the rest.Config object

func NewClientBackedDryRunGetterFromKubeconfig 

func NewClientBackedDryRunGetterFromKubeconfig(file string) (*ClientBackedDryRunGetter, error)

NewClientBackedDryRunGetterFromKubeconfig creates a new ClientBackedDryRunGetter instance from the given KubeConfig file

func (*ClientBackedDryRunGetter) Client 

func (clg *ClientBackedDryRunGetter) Client() clientset.Interface

Client gets the backing clientset.Interface

func (*ClientBackedDryRunGetter) HandleGetAction 

func (clg *ClientBackedDryRunGetter) HandleGetAction(action core.GetAction) (bool, runtime.Object, error)

HandleGetAction handles GET actions to the dryrun clientset this interface supports

func (*ClientBackedDryRunGetter) HandleListAction 

func (clg *ClientBackedDryRunGetter) HandleListAction(action core.ListAction) (bool, runtime.Object, error)

HandleListAction handles LIST actions to the dryrun clientset this interface supports

type DryRunClientOptions 

type DryRunClientOptions struct {
	Writer          io.Writer
	Getter          DryRunGetter
	PrependReactors []core.Reactor
	AppendReactors  []core.Reactor
	MarshalFunc     MarshalFunc
	PrintGETAndLIST bool
}

DryRunClientOptions specifies options to pass to NewDryRunClientWithOpts in order to get a dryrun clientset

func GetDefaultDryRunClientOptions added in v1.11.0 

func GetDefaultDryRunClientOptions(drg DryRunGetter, w io.Writer) DryRunClientOptions

GetDefaultDryRunClientOptions returns the default DryRunClientOptions values

type DryRunGetter 

type DryRunGetter interface {
	HandleGetAction(core.GetAction) (bool, runtime.Object, error)
	HandleListAction(core.ListAction) (bool, runtime.Object, error)
}

DryRunGetter is an interface that must be supplied to the NewDryRunClient function in order to contstruct a fully functional fake dryrun clientset

type InitDryRunGetter 

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

InitDryRunGetter implements the DryRunGetter interface and can be used to GET/LIST values in the dryrun fake clientset Need to handle these routes in a special manner: - GET /default/services/kubernetes -- must return a valid Service - GET /clusterrolebindings/system:nodes -- can safely return a NotFound error - GET /kube-system/secrets/bootstrap-token-* -- can safely return a NotFound error - GET /nodes/<node-name> -- must return a valid Node - ...all other, unknown GETs/LISTs will be logged

func NewInitDryRunGetter 

func NewInitDryRunGetter(controlPlaneName string, serviceSubnet string) *InitDryRunGetter

NewInitDryRunGetter creates a new instance of the InitDryRunGetter struct

func (*InitDryRunGetter) HandleGetAction 

func (idr *InitDryRunGetter) HandleGetAction(action core.GetAction) (bool, runtime.Object, error)

HandleGetAction handles GET actions to the dryrun clientset this interface supports

func (*InitDryRunGetter) HandleListAction 

func (idr *InitDryRunGetter) HandleListAction(action core.ListAction) (bool, runtime.Object, error)

HandleListAction handles GET actions to the dryrun clientset this interface supports. Currently there are no known LIST calls during kubeadm init this code has to take care of.

type KubeWaiter 

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

KubeWaiter is an implementation of Waiter that is backed by a Kubernetes client

func (*KubeWaiter) SetTimeout 

func (w *KubeWaiter) SetTimeout(timeout time.Duration)

SetTimeout adjusts the timeout to the specified duration

func (*KubeWaiter) WaitForAPI 

func (w *KubeWaiter) WaitForAPI() error

WaitForAPI waits for the API Server's /healthz endpoint to report "ok"

func (*KubeWaiter) WaitForHealthyKubelet 

func (w *KubeWaiter) WaitForHealthyKubelet(initalTimeout time.Duration, healthzEndpoint string) error

WaitForHealthyKubelet blocks until the kubelet /healthz endpoint returns 'ok'

func (*KubeWaiter) WaitForKubeletAndFunc added in v1.13.0 

func (w *KubeWaiter) WaitForKubeletAndFunc(f func() error) error

WaitForKubeletAndFunc waits primarily for the function f to execute, even though it might take some time. If that takes a long time, and the kubelet /healthz continuously are unhealthy, kubeadm will error out after a period of exponential backoff

func (*KubeWaiter) WaitForPodToDisappear 

func (w *KubeWaiter) WaitForPodToDisappear(podName string) error

WaitForPodToDisappear blocks until it timeouts or gets a "NotFound" response from the API Server when getting the Static Pod in question

func (*KubeWaiter) WaitForPodsWithLabel 

func (w *KubeWaiter) WaitForPodsWithLabel(kvLabel string) error

WaitForPodsWithLabel will lookup pods with the given label and wait until they are all reporting status as running.

func (*KubeWaiter) WaitForStaticPodControlPlaneHashes 

func (w *KubeWaiter) WaitForStaticPodControlPlaneHashes(nodeName string) (map[string]string, error)

WaitForStaticPodControlPlaneHashes blocks until it timeouts or gets a hash map for all components and their Static Pods

func (*KubeWaiter) WaitForStaticPodHashChange added in v1.10.2 

func (w *KubeWaiter) WaitForStaticPodHashChange(nodeName, component, previousHash string) error

WaitForStaticPodHashChange blocks until it timeouts or notices that the Mirror Pod (for the Static Pod, respectively) has changed This implicitly means this function blocks until the kubelet has restarted the Static Pod in question

func (*KubeWaiter) WaitForStaticPodSingleHash added in v1.9.0 

func (w *KubeWaiter) WaitForStaticPodSingleHash(nodeName string, component string) (string, error)

WaitForStaticPodSingleHash blocks until it timeouts or gets a hash for a single component and its Static Pod

type MarshalFunc 

type MarshalFunc func(runtime.Object, schema.GroupVersion) ([]byte, error)

MarshalFunc takes care of converting any object to a byte array for displaying the object to the user

type Waiter 

type Waiter interface {
	// WaitForAPI waits for the API Server's /healthz endpoint to become "ok"
	WaitForAPI() error
	// WaitForPodsWithLabel waits for Pods in the kube-system namespace to become Ready
	WaitForPodsWithLabel(kvLabel string) error
	// WaitForPodToDisappear waits for the given Pod in the kube-system namespace to be deleted
	WaitForPodToDisappear(staticPodName string) error
	// WaitForStaticPodSingleHash fetches sha256 hash for the control plane static pod
	WaitForStaticPodSingleHash(nodeName string, component string) (string, error)
	// WaitForStaticPodHashChange waits for the given static pod component's static pod hash to get updated.
	// By doing that we can be sure that the kubelet has restarted the given Static Pod
	WaitForStaticPodHashChange(nodeName, component, previousHash string) error
	// WaitForStaticPodControlPlaneHashes fetches sha256 hashes for the control plane static pods
	WaitForStaticPodControlPlaneHashes(nodeName string) (map[string]string, error)
	// WaitForHealthyKubelet blocks until the kubelet /healthz endpoint returns 'ok'
	WaitForHealthyKubelet(initalTimeout time.Duration, healthzEndpoint string) error
	// WaitForKubeletAndFunc is a wrapper for WaitForHealthyKubelet that also blocks for a function
	WaitForKubeletAndFunc(f func() error) error
	// SetTimeout adjusts the timeout to the specified duration
	SetTimeout(timeout time.Duration)
}

Waiter is an interface for waiting for criteria in Kubernetes to happen

func NewKubeWaiter 

func NewKubeWaiter(client clientset.Interface, timeout time.Duration, writer io.Writer) Waiter

NewKubeWaiter returns a new Waiter object that talks to the given Kubernetes cluster

Source Files

View all Source files

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.