--story=120070378 优化审计日志&执行后置初始化脚本 (merge request !2094)

Squash merge branch 'bcs-cluster-manager-1030' into 'master'
--story=120070378 优化审计日志&执行后置初始化脚本

 



TAPD: --story=120070378

bcs-services/bcs-cluster-manager/api/clustermanager/clustermanager.proto

@@ -12729,6 +12729,14 @@ message OperationLogDetail {
     title : "resourceName",
     description : "资源名称"
   }];
+  bool allowRetry = 11 [(grpc.gateway.protoc_gen_swagger.options.openapiv2_field) = {
+    title : "allowRetry",
+    description : "如果该操作是异步任务, 控制任务是否允许重试"
+  }];
+  bool allowSkip = 12 [(grpc.gateway.protoc_gen_swagger.options.openapiv2_field) = {
+    title : "allowSkip",
+    description : "如果该操作是异步任务, 控制任务是否允许跳过"
+  }];
 }
 
 message ListTaskStepLogsResponse {

bcs-services/bcs-cluster-manager/api/clustermanager/clustermanager.swagger.json

@@ -16874,6 +16874,18 @@
           "type": "string",
           "description": "资源名称",
           "title": "resourceName"
+        },
+        "allowRetry": {
+          "type": "boolean",
+          "format": "boolean",
+          "description": "如果该操作是异步任务, 控制任务是否允许重试",
+          "title": "allowRetry"
+        },
+        "allowSkip": {
+          "type": "boolean",
+          "format": "boolean",
+          "description": "如果该操作是异步任务, 控制任务是否允许跳过",
+          "title": "allowSkip"
         }
       }
     },

bcs-services/bcs-cluster-manager/internal/actions/operationlog/operationlog.go

@@ -279,6 +279,17 @@ func (ua *ListOperationLogsAction) appendTasks(taskIDs []string) error {
 			t.End = endTime
 
 			t.TaskName = autils.Translate(ua.ctx, t.TaskType, t.TaskName, "")
+			ua.resp.Data.Results[i].TaskType = t.TaskType
+
+			allowRetry := true
+			// attention: 开启CA节点自动扩缩容的任务不允许手动重试
+			if utils.SliceContainInString([]string{cloudprovider.UpdateNodeGroupDesiredNode.String(),
+				cloudprovider.CleanNodeGroupNodes.String()}, t.TaskType) &&
+				t.GetCommonParams()[cloudprovider.ManualKey.String()] != common.True {
+				allowRetry = false
+			}
+
+			ua.resp.Data.Results[i].AllowRetry = allowRetry
 			ua.resp.Data.Results[i].Message = autils.TranslateMsg(ua.ctx, v.ResourceType, v.TaskType, v.Message, t)
 			ua.resp.Data.Results[i].Task = t
 		}

bcs-services/bcs-cluster-manager/internal/actions/operationlog/taskrecord.go

@@ -83,9 +83,9 @@ func (ua *TaskRecordsAction) fetchTaskRecords() error {
 
 	allowRetry := true
 	// attention: 开启CA节点自动扩缩容不允许手动重试
-	if utils.StringContainInSlice(task.TaskType, []string{cloudprovider.UpdateNodeGroupDesiredNode.String(),
-		cloudprovider.CleanNodeGroupNodes.String()}) &&
-		task.GetCommonParams()[cloudprovider.ManualKey.String()] == common.False {
+	if utils.SliceContainInString([]string{cloudprovider.UpdateNodeGroupDesiredNode.String(),
+		cloudprovider.CleanNodeGroupNodes.String()}, task.TaskType) &&
+		task.GetCommonParams()[cloudprovider.ManualKey.String()] != common.True {
 		allowRetry = false
 	}
 

bcs-services/bcs-cluster-manager/internal/actions/task/update.go

@@ -156,6 +156,12 @@ func (ua *RetryAction) validate() error {
 	if err := ua.req.Validate(); err != nil {
 		return err
 	}
+
+	retry := allowTaskRetry(ua.task)
+	if !retry {
+		return fmt.Errorf("autoscaler task can't manual retry")
+	}
+
 	// check task status
 	switch ua.task.Status {
 	case cloudprovider.TaskStatusInit, cloudprovider.TaskStatusRunning, cloudprovider.TaskStatusSuccess:

bcs-services/bcs-cluster-manager/internal/actions/task/utils.go

@@ -24,8 +24,20 @@ import (
 	"github.com/Tencent/bk-bcs/bcs-services/bcs-cluster-manager/internal/cloudprovider"
 	"github.com/Tencent/bk-bcs/bcs-services/bcs-cluster-manager/internal/common"
 	"github.com/Tencent/bk-bcs/bcs-services/bcs-cluster-manager/internal/store"
+	"github.com/Tencent/bk-bcs/bcs-services/bcs-cluster-manager/internal/utils"
 )
 
+func allowTaskRetry(task *proto.Task) bool {
+	allowRetry := true
+	// attention: 开启CA节点自动扩缩容的任务不允许手动重试
+	if utils.SliceContainInString([]string{cloudprovider.UpdateNodeGroupDesiredNode.String(),
+		cloudprovider.CleanNodeGroupNodes.String()}, task.TaskType) && (task.GetCommonParams() != nil &&
+		task.GetCommonParams()[cloudprovider.ManualKey.String()] != common.True) {
+		allowRetry = false
+	}
+	return allowRetry
+}
+
 func updateTaskDataStatus(model store.ClusterManagerModel, task *proto.Task) error {
 	blog.Infof("updateTaskDataStatus[%s] taskType[%s]", task.TaskID, task.TaskType)
 

bcs-services/bcs-cluster-manager/internal/cloudprovider/common/job.go

@@ -15,6 +15,7 @@ package common
 import (
 	"context"
 	"fmt"
+	"strconv"
 	"strings"
 	"time"
 
@@ -207,7 +208,7 @@ func ExecuteScriptByJob(ctx context.Context, stepName, bizID, content string, ip
 		err   error
 	)
 
-	servers, err := GetIPCloudIDByNodeIPs(ctx, ips)
+	servers, err := GetIPCloudIDByNodeIPs(ctx, bizID, ips)
 	if err != nil {
 		blog.Errorf("task[%s] ExecuteScriptByJob failed: %v", taskID, err)
 		return "", err
@@ -272,20 +273,52 @@ func ExecuteScriptByJob(ctx context.Context, stepName, bizID, content string, ip
 	return job.GetJobTaskLink(jobID), nil
 }
 
+func getBizHosts(bizID string) (map[int64]cmdb.HostData, error) {
+	biz, err := strconv.Atoi(bizID)
+	if err != nil {
+		blog.Errorf("strconv BusinessID to int failed: %v", err)
+		return nil, err
+	}
+	hosts, err := cmdb.GetCmdbClient().FetchAllHostsByBizID(biz, false)
+	if err != nil {
+		blog.Errorf("cmdb FetchAllHostsByBizID failed: %v", err)
+		return nil, err
+	}
+
+	var (
+		hostsMap = make(map[int64]cmdb.HostData, 0)
+	)
+	for i := range hosts {
+		hostsMap[hosts[i].BKHostID] = hosts[i]
+	}
+	return hostsMap, nil
+}
+
 // GetIPCloudIDByNodeIPs get serverIP cloudInfo by cmdb
-func GetIPCloudIDByNodeIPs(ctx context.Context, ips []string) ([]job.ServerInfo, error) {
+func GetIPCloudIDByNodeIPs(ctx context.Context, bizID string, ips []string) ([]job.ServerInfo, error) {
 	taskID := cloudprovider.GetTaskIDFromContext(ctx)
 
+	// withoutBiz maybe get other biz host
 	hostDetailData, err := cmdb.GetCmdbClient().QueryAllHostInfoWithoutBiz(ips)
 	if err != nil {
 		blog.Errorf("task[%s] GetIPCloudIDByNodeIPs failed: %v", taskID, err)
 		return nil, err
 	}
 
+	bizHostsMap, err := getBizHosts(bizID)
+	if err != nil {
+		return nil, err
+	}
+
 	var (
 		servers = make([]job.ServerInfo, 0)
 	)
 	for _, host := range hostDetailData {
+		_, ok := bizHostsMap[host.BKHostID]
+		if !ok {
+			continue
+		}
+
 		servers = append(servers, job.ServerInfo{
 			BkCloudID: uint64(host.BkCloudID),
 			Ip:        host.BKHostInnerIP,

bcs-services/bcs-cluster-manager/internal/daemon/cluster_helath.go

@@ -80,14 +80,14 @@ func (d *Daemon) reportClusterHealthStatus(error chan<- error) {
 				error <- errLocal
 				return
 			}
-			_, err = kubeCli.Discovery().ServerVersion()
-			if err != nil {
+			_, errLocal = kubeCli.Discovery().ServerVersion()
+			if errLocal != nil {
 				blog.Errorf("reportClusterHealthStatus GetClusterClient failed: %v", errLocal)
 				// if options.GetEditionInfo().IsCommunicationEdition() {}
 				_ = d.updateClusterStatus(cls.ClusterID, common.StatusConnectClusterFailed)
 
 				metrics.ReportCloudClusterHealthStatus(cls.Provider, cls.ClusterID, 0)
-				error <- err
+				error <- errLocal
 				return
 			}