Terraform创建Google Cloud MongoDB告警策略失败求助

在Google Cloud Monitoring中，为Compute Engine实例上的MongoDB数据库创建告警策略时，通过GCP控制台可成功完成，但使用Terraform部署时持续失败，报错无法找到指定指标。

核心错误信息 #

执行terraform apply时收到的错误：

Error creating AlertPolicy: googleapi: Error 404: Cannot find metric(s) that match type = "workload.googleapis.com/mongodb.collection.count". If a metric was created recently, it could take up to 10 minutes to become available. Please try again soon

完整错误栈：

│ Error: Error creating AlertPolicy: googleapi: Error 404: Cannot find metric(s) that match type = "workload.googleapis.com/mongodb.collection.count". If a metric was created recently, it could take up to 10 minutes to become available. Please try again soon.
│ 
│   with module.monitor["XXX"].google_monitoring_alert_policy.monitor,
│   on monitor_wrapper/monitor.tf line 1, in resource "google_monitoring_alert_policy" "monitor":
│    1: resource "google_monitoring_alert_policy" "monitor" {

配置对比 #

GUI导出的可正常运行的告警策略 #

{  
  "displayName": "",  
  "userLabels": {},  
  "conditions": [    
    {      
      "displayName": "VM Instance - workload/mongodb.collection.count",      
      "conditionThreshold": {        
        "filter": "resource.type = \"gce_instance\" AND metric.type = \"workload.googleapis.com/mongodb.collection.count\"",        
        "aggregations": [          
          {            
            "alignmentPeriod": "600s",            
            "crossSeriesReducer": "REDUCE_NONE",            
            "perSeriesAligner": "ALIGN_MEAN"          
          }        
        ],        
        "comparison": "COMPARISON_GT",        
        "duration": "0s",        
        "trigger": {          
          "count": 1        
        }      
      }    
    }  
  ],  
  "alertStrategy": {    
    "autoClose": "604800s"  
  },  
  "combiner": "OR",  
  "enabled": true,  
  "notificationChannels": []
}

Terraform中部署失败的告警配置 #

mongodb_collections_critical = {  
  name     = "CRITICAL: VM Instance - MongoDB collection count > #threshold#"  
  combiner = "OR"

  thresholds = {    
    threshold_value = 30  
  }

  conditions = {    
    condition_1 = {      
      name = "CRITICAL: VM Instance - MongoDB collection count > #threshold#"      
      condition_threshold = {        
        comparison = "COMPARISON_GT"        
        duration   = "0s"        
        filter     = "resource.type = \"gce_instance\" AND metric.type = \"workload.googleapis.com/mongodb.collection.count\""

        aggregations = {          
          aggr1 = {            
            alignment_period   = "600s"            
            per_series_aligner = "ALIGN_MEAN"          
          }        
        }

        trigger_count = 1      
      }    
    }  
  }

  content = <<EOT            
    ALERT DOCUMENTATION          
  EOT

  pd_integration_level = "critical"    
}

已尝试的操作 #

• 确认指标数据正在生成，等待超过10分钟后重新执行部署
• 验证代码中metric.type的名称和语法，与GUI导出的配置完全一致
• 按照Google官方文档验证了Ops Agent的配置正确性

可能的解决方向 #

1. 补充聚合配置中的cross_series_reducer参数 #

GUI导出的策略中，aggregations明确包含"crossSeriesReducer": "REDUCE_NONE"，但你的Terraform配置缺失该参数。GCP监控API可能要求该字段必须存在，即使值为REDUCE_NONE，缺失会导致API无法识别指标的聚合规则，进而触发找不到指标的错误。

修改Terraform的聚合配置：

aggregations = {
  aggr1 = {
    alignment_period   = "600s"
    per_series_aligner = "ALIGN_MEAN"
    cross_series_reducer = "REDUCE_NONE" # 新增该行
  }
}

2. 检查模板占位符的渲染结果 #

配置中使用了#threshold#占位符，需确认这些占位符在Terraform模块中是否被正确替换为实际值。可通过terraform plan命令查看最终生成的告警策略配置，验证filter、name等字段是否完全符合预期，避免渲染异常导致的格式错误。

3. 升级Google Cloud Terraform Provider版本 #

旧版本的Terraform Provider可能对workload.googleapis.com系列指标的支持不完善，或API调用格式存在差异。尝试升级到最新稳定版的google provider：

terraform {
  required_providers {
    google = {
      source  = "hashicorp/google"
      version = ">= 4.70.0" # 替换为当前最新稳定版本
    }
  }
}

4. 验证权限与资源范围 #

• 确认Terraform使用的服务账号拥有monitoring.alertPolicies.create权限
• 确保Terraform操作的GCP项目与GUI中创建告警的项目一致
• 检查GCE实例的标签、区域等属性，确认filter中的resource.type = "gce_instance"能正确匹配目标实例，无额外范围限制导致指标无法被识别

内容的提问来源于stack exchange，提问作者LPrac