添加confirm确认显式指定分批数，保证幂等

.github/workflows/module_controller_ci_build_batch_deploy_to_aliyun.yml

@@ -167,7 +167,6 @@ jobs:
         run: |
           # 定义要等待的资源名称和字段值
           moduledeploymentname=$(kubectl get moduledeployment -o name)
-          desired_field_value=true
 
           # 定义等待的超时时间（以秒为单位）
           timeout_seconds=300
@@ -183,11 +182,11 @@ jobs:
             fi
 
             # 使用 kubectl get 命令获取资源对象的详细信息，并提取自定义字段的值
-            field_value=$(kubectl get $moduledeploymentname -o custom-columns=PAUSE:.spec.pause --no-headers)
+            field_value=$(kubectl get $moduledeploymentname -o custom-columns=CONFIRMBATCHNUM:.spec.confirmBatchNum --no-headers)
 
-            if [ "$field_value" == "$desired_field_value" ]; then
+            if [ "$field_value" == "0" ]; then
               echo "字段值已满足条件，执行分组确认"
-              kubectl patch $moduledeploymentname -p '{"spec":{"pause":false}}' --type=merge
+              kubectl patch $moduledeploymentname -p '{"spec":{"confirmBatchNum":1}}' --type=merge
               exit 0
             else
               echo "等待字段值满足条件..."
@@ -213,7 +212,6 @@ jobs:
         run: |
           # 定义要等待的资源名称和字段值
           moduledeploymentname=$(kubectl get moduledeployment -o name)
-          desired_field_value=true
 
           # 定义等待的超时时间（以秒为单位）
           timeout_seconds=300
@@ -229,11 +227,11 @@ jobs:
             fi
 
             # 使用 kubectl get 命令获取资源对象的详细信息，并提取自定义字段的值
-            field_value=$(kubectl get $moduledeploymentname -o custom-columns=PAUSE:.spec.pause --no-headers)
+            field_value=$(kubectl get $moduledeploymentname -o custom-columns=CONFIRMBATCHNUM:.spec.confirmBatchNum --no-headers)
 
-            if [ "$field_value" == "$desired_field_value" ]; then
+            if [ "$field_value" == "1" ]; then
               echo "字段值已满足条件，执行分组确认"
-              kubectl patch $moduledeploymentname -p '{"spec":{"pause":false}}' --type=merge
+              kubectl patch $moduledeploymentname -p '{"spec":{"confirmBatchNum":2}}' --type=merge
               exit 0
             else
               echo "等待字段值满足条件..."
@@ -353,7 +351,6 @@ jobs:
         run: |
           # 定义要等待的资源名称和字段值
           moduledeploymentname=$(kubectl get moduledeployment -o name)
-          desired_field_value=true
 
           # 定义等待的超时时间（以秒为单位）
           timeout_seconds=300
@@ -369,11 +366,11 @@ jobs:
             fi
 
             # 使用 kubectl get 命令获取资源对象的详细信息，并提取自定义字段的值
-            field_value=$(kubectl get $moduledeploymentname -o custom-columns=PAUSE:.spec.pause --no-headers)
+            field_value=$(kubectl get $moduledeploymentname -o custom-columns=CONFIRMBATCHNUM:.spec.confirmBatchNum --no-headers)
 
-            if [ "$field_value" == "$desired_field_value" ]; then
+            if [ "$field_value" == "0" ]; then
               echo "字段值已满足条件，执行分组确认"
-              kubectl patch $moduledeploymentname -p '{"spec":{"pause":false}}' --type=merge
+              kubectl patch $moduledeploymentname -p '{"spec":{"confirmBatchNum":1}}' --type=merge
               exit 0
             else
               echo "等待字段值满足条件..."
@@ -400,7 +397,6 @@ jobs:
         run: |
           # 定义要等待的资源名称和字段值
           moduledeploymentname=$(kubectl get moduledeployment -o name)
-          desired_field_value=true
 
           # 定义等待的超时时间（以秒为单位）
           timeout_seconds=300
@@ -416,11 +412,11 @@ jobs:
             fi
 
             # 使用 kubectl get 命令获取资源对象的详细信息，并提取自定义字段的值
-            field_value=$(kubectl get $moduledeploymentname -o custom-columns=PAUSE:.spec.pause --no-headers)
+            field_value=$(kubectl get $moduledeploymentname -o custom-columns=CONFIRMBATCHNUM:.spec.confirmBatchNum --no-headers)
 
-            if [ "$field_value" == "$desired_field_value" ]; then
+            if [ "$field_value" == "1" ]; then
               echo "字段值已满足条件，执行分组确认"
-              kubectl patch $moduledeploymentname -p '{"spec":{"pause":false}}' --type=merge
+              kubectl patch $moduledeploymentname -p '{"spec":{"confirmBatchNum":2}}' --type=merge
               exit 0
             else
               echo "等待字段值满足条件..."
@@ -496,7 +492,6 @@ jobs:
         run: |
           # 定义要等待的资源名称和字段值
           moduledeploymentname=$(kubectl get moduledeployment -o name)
-          desired_field_value=true
 
           # 定义等待的超时时间（以秒为单位）
           timeout_seconds=300
@@ -512,11 +507,11 @@ jobs:
             fi
 
             # 使用 kubectl get 命令获取资源对象的详细信息，并提取自定义字段的值
-            field_value=$(kubectl get $moduledeploymentname -o custom-columns=PAUSE:.spec.pause --no-headers)
+            field_value=$(kubectl get $moduledeploymentname -o custom-columns=CONFIRMBATCHNUM:.spec.confirmBatchNum --no-headers)
 
-            if [ "$field_value" == "$desired_field_value" ]; then
+            if [ "$field_value" == "0" ]; then
               echo "字段值已满足条件，执行分组确认"
-              kubectl patch $moduledeploymentname -p '{"spec":{"pause":false}}' --type=merge
+              kubectl patch $moduledeploymentname -p '{"spec":{"confirmBatchNum":1}}' --type=merge
               exit 0
             else
               echo "等待字段值满足条件..."
@@ -579,7 +574,6 @@ jobs:
         run: |
           # 定义要等待的资源名称和字段值
           moduledeploymentname=$(kubectl get moduledeployment -o name)
-          desired_field_value=true
 
           # 定义等待的超时时间（以秒为单位）
           timeout_seconds=300
@@ -595,11 +589,11 @@ jobs:
             fi
 
             # 使用 kubectl get 命令获取资源对象的详细信息，并提取自定义字段的值
-            field_value=$(kubectl get $moduledeploymentname -o custom-columns=PAUSE:.spec.pause --no-headers)
+            field_value=$(kubectl get $moduledeploymentname -o custom-columns=CONFIRMBATCHNUM:.spec.confirmBatchNum --no-headers)
 
-            if [ "$field_value" == "$desired_field_value" ]; then
+            if [ "$field_value" == "1" ]; then
               echo "字段值已满足条件，执行分组确认"
-              kubectl patch $moduledeploymentname -p '{"spec":{"pause":false}}' --type=merge
+              kubectl patch $moduledeploymentname -p '{"spec":{"confirmBatchNum":2}}' --type=merge
               exit 0
             else
               echo "等待字段值满足条件..."

.github/workflows/module_controller_ci_build_batch_scaleup_then_scaledown_deploy_to_aliyun.yml

@@ -167,7 +167,6 @@ jobs:
         run: |
           # 定义要等待的资源名称和字段值
           moduledeploymentname=$(kubectl get moduledeployment -o name)
-          desired_field_value=true
 
           # 定义等待的超时时间（以秒为单位）
           timeout_seconds=300
@@ -183,11 +182,11 @@ jobs:
             fi
 
             # 使用 kubectl get 命令获取资源对象的详细信息，并提取自定义字段的值
-            field_value=$(kubectl get $moduledeploymentname -o custom-columns=PAUSE:.spec.pause --no-headers)
+            field_value=$(kubectl get $moduledeploymentname -o custom-columns=CONFIRMBATCHNUM:.spec.confirmBatchNum --no-headers)
 
-            if [ "$field_value" == "$desired_field_value" ]; then
+            if [ "$field_value" == "0" ]; then
               echo "字段值已满足条件，执行分组确认"
-              kubectl patch $moduledeploymentname -p '{"spec":{"pause":false}}' --type=merge
+              kubectl patch $moduledeploymentname -p '{"spec":{"confirmBatchNum":1}}' --type=merge
               exit 0
             else
               echo "等待字段值满足条件..."
@@ -579,7 +578,6 @@ jobs:
         run: |
           # 定义要等待的资源名称和字段值
           moduledeploymentname=$(kubectl get moduledeployment -o name)
-          desired_field_value=true
 
           # 定义等待的超时时间（以秒为单位）
           timeout_seconds=300
@@ -595,11 +593,11 @@ jobs:
             fi
 
             # 使用 kubectl get 命令获取资源对象的详细信息，并提取自定义字段的值
-            field_value=$(kubectl get $moduledeploymentname -o custom-columns=PAUSE:.spec.pause --no-headers)
+            field_value=$(kubectl get $moduledeploymentname -o custom-columns=CONFIRMBATCHNUM:.spec.confirmBatchNum --no-headers)
 
-            if [ "$field_value" == "$desired_field_value" ]; then
+            if [ "$field_value" == "0" ]; then
               echo "字段值已满足条件，执行分组确认"
-              kubectl patch $moduledeploymentname -p '{"spec":{"pause":false}}' --type=merge
+              kubectl patch $moduledeploymentname -p '{"spec":{"confirmBatchNum":1}}' --type=merge
               exit 0
             else
               echo "等待字段值满足条件..."
@@ -738,7 +736,6 @@ jobs:
         run: |
           # 定义要等待的资源名称和字段值
           moduledeploymentname=$(kubectl get moduledeployment -o name)
-          desired_field_value=true
 
           # 定义等待的超时时间（以秒为单位）
           timeout_seconds=300
@@ -754,11 +751,11 @@ jobs:
             fi
 
             # 使用 kubectl get 命令获取资源对象的详细信息，并提取自定义字段的值
-            field_value=$(kubectl get $moduledeploymentname -o custom-columns=PAUSE:.spec.pause --no-headers)
+            field_value=$(kubectl get $moduledeploymentname -o custom-columns=CONFIRMBATCHNUM:.spec.confirmBatchNum --no-headers)
 
-            if [ "$field_value" == "$desired_field_value" ]; then
+            if [ "$field_value" == "0" ]; then
               echo "字段值已满足条件，执行分组确认"
-              kubectl patch $moduledeploymentname -p '{"spec":{"pause":false}}' --type=merge
+              kubectl patch $moduledeploymentname -p '{"spec":{"confirmBatchNum":1}}' --type=merge
               exit 0
             else
               echo "等待字段值满足条件..."

.github/workflows/module_controller_ci_build_batch_symmetric_deploy_to_aliyun.yml

@@ -167,7 +167,6 @@ jobs:
         run: |
           # 定义要等待的资源名称和字段值
           moduledeploymentname=$(kubectl get moduledeployment -o name)
-          desired_field_value=true
 
           # 定义等待的超时时间（以秒为单位）
           timeout_seconds=300
@@ -183,11 +182,11 @@ jobs:
             fi
 
             # 使用 kubectl get 命令获取资源对象的详细信息，并提取自定义字段的值
-            field_value=$(kubectl get $moduledeploymentname -o custom-columns=PAUSE:.spec.pause --no-headers)
+            field_value=$(kubectl get $moduledeploymentname -o custom-columns=CONFIRMBATCHNUM:.spec.confirmBatchNum --no-headers)
 
-            if [ "$field_value" == "$desired_field_value" ]; then
+            if [ "$field_value" == "0" ]; then
               echo "字段值已满足条件，执行分组确认"
-              kubectl patch $moduledeploymentname -p '{"spec":{"pause":false}}' --type=merge
+              kubectl patch $moduledeploymentname -p '{"spec":{"confirmBatchNum":1}}' --type=merge
               exit 0
             else
               echo "等待字段值满足条件..."
@@ -213,7 +212,6 @@ jobs:
         run: |
           # 定义要等待的资源名称和字段值
           moduledeploymentname=$(kubectl get moduledeployment -o name)
-          desired_field_value=true
 
           # 定义等待的超时时间（以秒为单位）
           timeout_seconds=300
@@ -229,11 +227,11 @@ jobs:
             fi
 
             # 使用 kubectl get 命令获取资源对象的详细信息，并提取自定义字段的值
-            field_value=$(kubectl get $moduledeploymentname -o custom-columns=PAUSE:.spec.pause --no-headers)
+            field_value=$(kubectl get $moduledeploymentname -o custom-columns=CONFIRMBATCHNUM:.spec.confirmBatchNum --no-headers)
 
-            if [ "$field_value" == "$desired_field_value" ]; then
+            if [ "$field_value" == "1" ]; then
               echo "字段值已满足条件，执行分组确认"
-              kubectl patch $moduledeploymentname -p '{"spec":{"pause":false}}' --type=merge
+              kubectl patch $moduledeploymentname -p '{"spec":{"confirmBatchNum":2}}' --type=merge
               exit 0
             else
               echo "等待字段值满足条件..."

module-controller/api/v1alpha1/moduledeployment_types.go

@@ -163,10 +163,9 @@ type ModuleDeploymentSpec struct {
 
 	ProgressDeadlineSeconds int32 `json:"progressDeadlineSeconds,omitempty"`
 
-	// Indicates that the moduleDeployment is paused and will not be processed by the
-	// moduleDeployment controller.
-	// +optional
-	Pause bool `json:"pause,omitempty"`
+	// +kubebuilder:default:=0
+	// +kubebuilder:validation:Minimum=0
+	ConfirmBatchNum int32 `json:"confirmBatchNum,omitempty"`
 
 	OperationStrategy ModuleOperationStrategy `json:"operationStrategy,omitempty"`
 

module-controller/config/crd/bases/serverless.alipay.com_moduledeployments.yaml

@@ -41,6 +41,11 @@ spec:
                   Important: Run "make" to regenerate code after modifying this file'
                 minLength: 1
                 type: string
+              confirmBatchNum:
+                default: 0
+                format: int32
+                minimum: 0
+                type: integer
               minReadySeconds:
                 format: int32
                 type: integer
@@ -75,10 +80,6 @@ spec:
                   useBeta:
                     type: boolean
                 type: object
-              pause:
-                description: Indicates that the moduleDeployment is paused and will
-                  not be processed by the moduleDeployment controller.
-                type: boolean
               progressDeadlineSeconds:
                 format: int32
                 type: integer

module-controller/internal/controller/moduledeployment_controller.go

@@ -107,7 +107,10 @@ func (r *ModuleDeploymentReconciler) Reconcile(ctx context.Context, req ctrl.Req
 		event.PublishModuleDeploymentCreateEvent(r.Client, ctx, moduleDeployment)
 	}
 
-	if moduleDeployment.Spec.Pause {
+	if moduleDeployment.Status.ReleaseStatus != nil &&
+		moduleDeployment.Spec.ConfirmBatchNum < moduleDeployment.Status.ReleaseStatus.CurrentBatch &&
+		moduleDeployment.Status.ReleaseStatus.Progress == v1alpha1.ModuleDeploymentReleaseProgressPaused {
+
 		return ctrl.Result{}, nil
 	}
 
@@ -162,19 +165,26 @@ func (r *ModuleDeploymentReconciler) Reconcile(ctx context.Context, req ctrl.Req
 				return ctrl.Result{}, err
 			}
 		}
-	case v1alpha1.ModuleDeploymentReleaseProgressWaitingForConfirmation:
-		moduleDeployment.Spec.Pause = true
-		if err := r.Update(ctx, moduleDeployment); err != nil {
-			return ctrl.Result{}, err
+
+		if moduleDeployment.Spec.ConfirmBatchNum > 0 {
+			moduleDeployment.Spec.ConfirmBatchNum = 0
+			if err := utils.UpdateResource(r.Client, ctx, moduleDeployment); err != nil {
+				return ctrl.Result{}, err
+			}
 		}
 
+	case v1alpha1.ModuleDeploymentReleaseProgressWaitingForConfirmation:
+
 		moduleDeployment.Status.ReleaseStatus.Progress = v1alpha1.ModuleDeploymentReleaseProgressPaused
 		log.Log.Info("update moduleDeployment releaseStatus progress to paused", "moduleDeploymentName", moduleDeployment.Name)
 		if err := utils.UpdateStatus(r.Client, ctx, moduleDeployment); err != nil {
 			return ctrl.Result{}, utils.Error(err, "update moduleDeployment releaseStatus progress to paused failed")
 		}
 	case v1alpha1.ModuleDeploymentReleaseProgressPaused:
-		if !moduleDeployment.Spec.Pause && time.Since(moduleDeployment.Status.ReleaseStatus.NextReconcileTime.Time) >= 0 {
+		if time.Since(moduleDeployment.Status.ReleaseStatus.NextReconcileTime.Time) >= 0 &&
+			moduleDeployment.Spec.ConfirmBatchNum == moduleDeployment.Status.ReleaseStatus.CurrentBatch {
+
+			moduleDeployment.Status.ReleaseStatus.CurrentBatch += 1
 			moduleDeployment.Status.ReleaseStatus.Progress = v1alpha1.ModuleDeploymentReleaseProgressExecuting
 			log.Log.Info("update moduleDeployment progress from paused to executing", "moduleDeploymentName", moduleDeployment.Name)
 			if err := utils.UpdateStatus(r.Client, ctx, moduleDeployment); err != nil {
@@ -465,7 +475,6 @@ func (r *ModuleDeploymentReconciler) updateModuleReplicaSet(ctx context.Context,
 		}
 	}
 	// TODO update current batch
-	moduleDeployment.Status.ReleaseStatus.CurrentBatch += 1
 	moduleDeployment.Status.ReleaseStatus.BatchProgress = v1alpha1.ModuleDeploymentReleaseProgressExecuting
 	log.Log.Info("update moduleDeployment batch progress to executing", "moduleDeploymentName", moduleDeployment.Name)
 	err = utils.UpdateStatus(r.Client, ctx, moduleDeployment)