如何用Terraform嵌套循环批量部署多环境Azure Monitor Alert

步骤1：重构locals.tf，结构化环境与告警配置 #

首先将重复的环境列表和告警模板抽离到locals.tf，通过嵌套for循环生成环境+告警的唯一组合映射，为后续批量迭代提供数据源：

# locals.tf
locals {
  # 定义所有需要部署的环境，新增/删除环境只需修改这里
  environments = ["sandbox", "test", "prod"]

  # 通用告警模板：将重复的告警配置统一管理，支持差异化调整
  alert_templates = {
    high_cpu_usage = {
      display_name = "High CPU Usage Alert"
      severity     = 2
      criteria = {
        metric_namespace     = "Microsoft.Compute/virtualMachines"
        metric_name          = "Percentage CPU"
        operator             = "GreaterThan"
        # 支持不同环境设置不同阈值（可选）
        threshold            = {
          sandbox = 90
          test    = 85
          prod    = 80
        }
        aggregation          = "Average"
        window_size          = "PT5M"
        evaluation_frequency = "PT1M"
      }
      action_group_id = "/subscriptions/your-sub-id/resourceGroups/your-rg/providers/Microsoft.Insights/actionGroups/your-action-group"
    }
    low_disk_space = {
      display_name = "Low Disk Space Alert"
      severity     = 3
      criteria = {
        metric_namespace     = "Microsoft.Compute/virtualMachines"
        metric_name          = "FreeSpacePercentage"
        operator             = "LessThan"
        threshold            = 20 # 所有环境共用阈值
        aggregation          = "Average"
        window_size          = "PT15M"
        evaluation_frequency = "PT5M"
      }
      action_group_id = "/subscriptions/your-sub-id/resourceGroups/your-rg/providers/Microsoft.Insights/actionGroups/your-action-group"
    }
    # 可继续添加更多告警模板
  }

  # 生成环境与告警的组合映射：用嵌套for循环遍历所有环境+告警，生成唯一键
  environment_alert_pairs = merge([
    for env in local.environments : {
      for alert_key, alert_cfg in local.alert_templates : "${env}-${alert_key}" => {
        env_name = env
        # 处理差异化阈值（如果有），否则直接复用模板配置
        alert_config = merge(alert_cfg, {
          criteria = merge(alert_cfg.criteria, {
            threshold = try(alert_cfg.criteria.threshold[env], alert_cfg.criteria.threshold)
          })
        })
      }
    }
  ]...)
}

步骤2：在monitor.tf中用for_each批量调用告警模块 #

通过for_each遍历上面生成的environment_alert_pairs，为每个环境的每个告警创建实例：

# monitor.tf
module "azure_monitor_alerts" {
  source = "./modules/azure-monitor-alert" # 替换为你的Azure Monitor告警模块路径

  # 遍历所有环境+告警的组合，每个组合对应一个告警资源实例
  for_each = local.environment_alert_pairs

  # 传递环境标识：用于区分不同环境的资源（比如资源命名、资源组）
  environment = each.value.env_name
  resource_group_name = "${each.value.env_name}-monitor-rg" # 假设每个环境有独立资源组

  # 传递告警核心配置
  alert_display_name = "${each.value.env_name} - ${each.value.alert_config.display_name}"
  alert_severity     = each.value.alert_config.severity
  metric_criteria    = each.value.alert_config.criteria
  action_group_id    = each.value.alert_config.action_group_id

  # 其他模块必填参数（根据你的模块定义补充）
  enabled = true
}

关键说明 #

1. 唯一键生成：${env}-${alert_key}确保每个告警实例的键唯一，Terraform会以此为标识管理资源生命周期。
2. 差异化配置：通过try和merge函数，可以轻松实现部分告警在不同环境下的参数差异（比如阈值、动作组），无需重复编写配置块。
3. 扩展性：新增环境只需在environments列表中添加；新增告警只需在alert_templates中添加模板，无需修改模块调用代码。

内容的提问来源于stack exchange，提问作者hitman126