开始
mkdir deploy-scaler
cd deploy-scaler
go mod init deploy-scaler
kubebuilder init -domain scaler.com
kubebuilder create api -kind Scaler --version vl -group api
编写 api/v1/scaler_types.go
package v1
import (
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
)
const (
SCALED = "Scaled"
FAILED = "Failed"
PENDING = "Pending"
RESTORED = "Restored"
)
// EDIT THIS FILE! THIS IS SCAFFOLDING FOR YOU TO OWN!
// NOTE: json tags are required. Any new fields you add must have json tags for the fields to be serialized.
type DeploymentInfo struct {
Replicas int32 `json:"replicas"`
Namespace string `json:"namespace"`
}
type NamespacedName struct {
Name string `json:"name"`
Namespace string `json:"namespace"`
}
// ScalerSpec defines the desired state of Scaler
type ScalerSpec struct {
// INSERT ADDITIONAL SPEC FIELDS - desired state of cluster
// Important: Run "make" to regenerate code after modifying this file
// +kubebuilder:validation:Maximum=23
// +kubebuilder:validation:Minimum=0
// +kubebuilder:validation:Required
Start int `json:"start"`
// +kubebuilder:validation:Maximum=24
// +kubebuilder:validation:Minimum=0
// +kubebuilder:validation:Required
End int `json:"end"`
Replicas int32 `json:"replicas"`
Deployments []NamespacedName `json:"deployments"`
}
// ScalerStatus defines the observed state of Scaler
type ScalerStatus struct {
// INSERT ADDITIONAL STATUS FIELD - define observed state of cluster
// Important: Run "make" to regenerate code after modifying this file
Status string `json:"status"`
}
//+kubebuilder:object:root=true
//+kubebuilder:subresource:status
//+kubebuilder:printcolumn:name="Status",type=string,JSONPath=`.status.status`
//+kubebuilder:printcolumn:name="Age",type="date",JSONPath=`.metadata.creationTimestamp`
// Scaler is the Schema for the scalers API
type Scaler struct {
metav1.TypeMeta `json:",inline"`
metav1.ObjectMeta `json:"metadata,omitempty"`
Spec ScalerSpec `json:"spec,omitempty"`
Status ScalerStatus `json:"status,omitempty"`
}
//+kubebuilder:object:root=true
// ScalerList contains a list of Scaler
type ScalerList struct {
metav1.TypeMeta `json:",inline"`
metav1.ListMeta `json:"metadata,omitempty"`
Items []Scaler `json:"items"`
}
func init() {
SchemeBuilder.Register(&Scaler{}, &ScalerList{})
}
编写 controller
package controllers
import (
"context"
"fmt"
v1 "k8s.io/api/apps/v1"
"k8s.io/apimachinery/pkg/types"
"k8s.io/apimachinery/pkg/util/json"
"sigs.k8s.io/controller-runtime/pkg/controller/controllerutil"
"sigs.k8s.io/controller-runtime/pkg/log"
"time"
apiv1alpha1 "github.com/jack410/deploy-scaler/api/v1alpha1"
"k8s.io/apimachinery/pkg/runtime"
ctrl "sigs.k8s.io/controller-runtime"
"sigs.k8s.io/controller-runtime/pkg/client"
)
const finalizer = "scalers.api.scaler.com/finalizer"
var logger = log.Log.WithName("scaler_controller")
var originalDeploymentInfo = make(map[string]apiv1alpha1.DeploymentInfo)
var annotations = make(map[string]string)
// ScalerReconciler reconciles a Scaler object
type ScalerReconciler struct {
client.Client
Scheme *runtime.Scheme
}
//+kubebuilder:rbac:groups=api.scaler.com,resources=scalers,verbs=get;list;watch;create;update;patch;delete
//+kubebuilder:rbac:groups=api.scaler.com,resources=scalers/status,verbs=get;update;patch
//+kubebuilder:rbac:groups=api.scaler.com,resources=scalers/finalizers,verbs=update
// Reconcile is part of the main kubernetes reconciliation loop which aims to
// move the current state of the cluster closer to the desired state.
// TODO(user): Modify the Reconcile function to compare the state specified by
// the Scaler object against the actual cluster state, and then
// perform operations to make the cluster state reflect the state specified by
// the user.
//
// For more details, check Reconcile and its Result here:
// - https://pkg.go.dev/sigs.k8s.io/controller-runtime@v0.12.1/pkg/reconcile
func (r *ScalerReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctrl.Result, error) {
log := logger.WithValues("Request.Namespace", req.Namespace, "Request.Name", req.Name)
log.Info("Reconcile called")
//创建一个scaler实例
scaler := &apiv1alpha1.Scaler{}
err := r.Get(ctx, req.NamespacedName, scaler)
if err != nil {
//如果没有发现这个scaler实例，我们就用 client.IgnoreNotFound(err)来忽略错误，让进程不中断
return ctrl.Result{}, client.IgnoreNotFound(err)
}
if scaler.ObjectMeta.DeletionTimestamp.IsZero() {
if !controllerutil.ContainsFinalizer(scaler, finalizer) {
controllerutil.AddFinalizer(scaler, finalizer)
log.Info("add finalizer.")
err := r.Update(ctx, scaler)
if err != nil {
return ctrl.Result{}, client.IgnoreNotFound(err)
}
}
if scaler.Status.Status == "" {
scaler.Status.Status = apiv1alpha1.PENDING
err := r.Status().Update(ctx, scaler)
if err != nil {
return ctrl.Result{}, client.IgnoreNotFound(err)
}
//将scaler中管理的deployments的副本数和namespace名称加到annotations里
if err := addAnnotations(scaler, r, ctx); err != nil {
return ctrl.Result{}, client.IgnoreNotFound(err)
}
}
//开始执行scaler的逻辑
startTime := scaler.Spec.Start
endTime := scaler.Spec.End
replicas := scaler.Spec.Replicas
currenHour := time.Now().Local().Hour()
log.Info(fmt.Sprintf("currentTIme: %d", currenHour))
//从startime开始endtime为止
if currenHour >= startTime && currenHour < endTime {
if scaler.Status.Status != apiv1alpha1.SCALED {
log.Info("starting to call scaleDeployment func.")
err := scaleDeployment(scaler, r, ctx, replicas)
if err != nil {
return ctrl.Result{}, err
}
}
} else {
if scaler.Status.Status == apiv1alpha1.SCALED {
restoreDeployment(scaler, r, ctx)
}
}
} else {
log.Info("starting deletion flow.")
if scaler.Status.Status == apiv1alpha1.SCALED {
log.Info("restore Deployment.")
err := restoreDeployment(scaler, r, ctx)
if err != nil {
return ctrl.Result{}, err
}
log.Info("remove finalizer.")
controllerutil.RemoveFinalizer(scaler, finalizer)
err = r.Update(ctx, scaler)
if err != nil {
return ctrl.Result{}, err
}
log.Info("finalizer removed.")
} else {
controllerutil.RemoveFinalizer(scaler, finalizer)
err = r.Update(ctx, scaler)
if err != nil {
return ctrl.Result{}, err
}
log.Info("finalizer removed.")
}
log.Info("scaler removed.")
}
return ctrl.Result{RequeueAfter: time.Duration(10 * time.Second)}, nil
}
func restoreDeployment(scaler *apiv1alpha1.Scaler, r *ScalerReconciler, ctx context.Context) error {
logger.Info("starting to return to the original state")
for name, originalDeployInfo := range originalDeploymentInfo {
deployment := &v1.Deployment{}
if err := r.Get(ctx, types.NamespacedName{
Name: name,
Namespace: originalDeployInfo.Namespace,
}, deployment); err != nil {
return err
}
if deployment.Spec.Replicas != &originalDeployInfo.Replicas {
deployment.Spec.Replicas = &originalDeployInfo.Replicas
if err := r.Update(ctx, deployment); err != nil {
return err
}
}
}
scaler.Status.Status = apiv1alpha1.RESTORED
err := r.Status().Update(ctx, scaler)
if err != nil {
return err
}
return nil
}
func scaleDeployment(scaler *apiv1alpha1.Scaler, r *ScalerReconciler, ctx context.Context, replicas int32) error {
//从scaler实例中遍历deployments
for _, deploy := range scaler.Spec.Deployments {
//创建一个新的deployment实例
deployment := &v1.Deployment{}
err := r.Get(ctx, types.NamespacedName{
Name: deploy.Name,
Namespace: deploy.Namespace,
}, deployment)
if err != nil {
return err
}
//判断当前k8s集群中的deployment副本数是否等于scaler中指定的副本数
if deployment.Spec.Replicas != &replicas {
deployment.Spec.Replicas = &replicas
err := r.Update(ctx, deployment)
if err != nil {
scaler.Status.Status = apiv1alpha1.FAILED
r.Status().Update(ctx, scaler)
return err
}
scaler.Status.Status = apiv1alpha1.SCALED
r.Status().Update(ctx, scaler)
}
}
return nil
}
func addAnnotations(scaler *apiv1alpha1.Scaler, r *ScalerReconciler, ctx context.Context) error {
//记录deployments的原始副本数和namespace名称
for _, deploy := range scaler.Spec.Deployments {
deployment := &v1.Deployment{}
if err := r.Get(ctx, types.NamespacedName{
Name: deploy.Name,
Namespace: deploy.Namespace,
}, deployment); err != nil {
return err
}
//开始记录
if *deployment.Spec.Replicas != scaler.Spec.Replicas {
logger.Info("add original state to originalDeploymentInfo map.")
originalDeploymentInfo[deployment.Name] = apiv1alpha1.DeploymentInfo{
Namespace: deployment.Namespace,
Replicas: *deployment.Spec.Replicas,
}
}
}
//将原始记录加到annotations里
for deploymentName, info := range originalDeploymentInfo {
//将info转换为json
infoJson, err := json.Marshal(info)
if err != nil {
return err
}
//将infoJson存到annotations map里
annotations[deploymentName] = string(infoJson)
}
//更新scaler的annotations
scaler.ObjectMeta.Annotations = annotations
err := r.Update(ctx, scaler)
if err != nil {
return err
}
return nil
}
// SetupWithManager sets up the controller with the Manager.
func (r *ScalerReconciler) SetupWithManager(mgr ctrl.Manager) error {
return ctrl.NewControllerManagedBy(mgr).
For(&apiv1alpha1.Scaler{}).
Complete(r)
}
执行 make manifests
修改 config/samples/下的系统默认的yaml,增加spec的修改
name: deploy1 就是要扩缩容的deployment，根据自己实际情况
apiVersion: api.scaler.com/v1
kind: Scaler
metadata:
name: scaler-sample
spec:
start: 20
end: 21
replicas: 2
deployments:
- name: deploy1
namespace: default
- name: deploy2
namespace: default
安装crd
make install
可以用 kubectl get api-resources 查询
运行
make run
另一个窗口运行
kubectl apply -f config/samples/
