Cherry-pick commit d103fd3

builder/deploy/deployer.go

@@ -1047,13 +1047,19 @@ func (d *deployer) CheckResourceAvailable(ctx context.Context, clusterId string,
 	}
 
 	if clusterResources.Status == types.ClusterStatusUnavailable {
-		return false, fmt.Errorf("failed to check cluster available resource due to cluster %s status is %s",
+		err := fmt.Errorf("failed to check cluster available resource due to cluster %s status is %s",
 			clusterId, clusterResources.Status)
+		return false, errorx.ClusterUnavailable(err, errorx.Ctx().
+			Set("cluster ID", clusterId).
+			Set("region", clusterResources.Region))
 	}
 
 	if clusterResources.ResourceStatus != types.StatusUncertain && !CheckResource(clusterResources, hardWare) {
-		return false, fmt.Errorf("required resource on cluster %s is not enough with resource status %s",
+		err := fmt.Errorf("required resource on cluster %s is not enough with resource status %s",
 			clusterId, clusterResources.ResourceStatus)
+		return false, errorx.NotEnoughResource(err, errorx.Ctx().
+			Set("cluster ID", clusterId).
+			Set("region", clusterResources.Region))
 	}
 
 	return true, nil

common/errorx/error_task.go

@@ -7,6 +7,8 @@ const (
 	multiHostInferenceNotSupported
 	multiHostInferenceReplicaCount
 	multiHostNotebookNotSupported
+	notEnoughResource
+	clusterUnavailable
 )
 
 var (
@@ -61,6 +63,30 @@ var (
 	//
 	// zh-HK: 多主機推理僅支持大於 0 的最低副本數
 	ErrMultiHostInferenceReplicaCount = CustomError{prefix: errTaskPrefix, code: multiHostInferenceReplicaCount}
+	// not enough resource to run the task
+	//
+	// Description: The task requires more resources than are available in the cluster. This error occurs when the cluster does not have sufficient capacity to run the task.
+	//
+	// Description_ZH: 任务需要的资源超过了集群可用的资源。当集群资源不足时，会出现此错误。
+	//
+	// en-US: Not enough resource to run the task
+	//
+	// zh-CN: 集群资源不足
+	//
+	// zh-HK: 集群資源不足
+	ErrNotEnoughResource = CustomError{prefix: errTaskPrefix, code: notEnoughResource}
+	// cluster is unavailable to run the task
+	//
+	// Description: The cluster is currently unavailable, either due to maintenance or other reasons. This error occurs when the cluster is not ready to accept new tasks.
+	//
+	// Description_ZH: 集群当前不可用，可能是由于维护或其他原因。当集群未准备好接受新任务时，会出现此错误。
+	//
+	// en-US: Cluster is unavailable to run the task
+	//
+	// zh-CN: 集群当前不可用
+	//
+	// zh-HK: 集群當前不可用
+	ErrClusterUnavailable = CustomError{prefix: errTaskPrefix, code: clusterUnavailable}
 )
 
 func NoEntryFile(err error, ctx context) error {
@@ -71,3 +97,21 @@ func NoEntryFile(err error, ctx context) error {
 		code:    noEntryFile,
 	}
 }
+
+func NotEnoughResource(err error, ctx context) error {
+	return CustomError{
+		prefix:  errTaskPrefix,
+		context: ctx,
+		err:     err,
+		code:    notEnoughResource,
+	}
+}
+
+func ClusterUnavailable(err error, ctx context) error {
+	return CustomError{
+		prefix:  errTaskPrefix,
+		context: ctx,
+		err:     err,
+		code:    clusterUnavailable,
+	}
+}