如何在Terraform中用单个资源实现多文件上传至多个S3存储桶

该需求完全可以在Terraform中实现，核心思路是先将三个按索引对应的配置项配对，再展平所有待上传文件的映射关系作为for_each的输入，无需拆分多个资源块。

实现逻辑 #

1. 按索引配对source_file_path、bucket_name、data_folder三个列表的对应项，保证source1对应bucket1、path1的映射关系
2. 遍历每个配对项，调用fileset获取该目录下所有待上传文件，最终展平为全局唯一键的映射，供资源块的for_each使用
3. 资源块直接读取映射内的参数完成上传配置

完整代码示例 #

# 原有变量定义，新增校验规则保证三个列表长度一致
variable "source_file_path"{
   type = list(string)
   description = "Path from where objects are to be uploaded"
   validation {
     condition = length(var.source_file_path) == length(var.bucket_name) && length(var.source_file_path) == length(var.data_folder)
     error_message = "source_file_path、bucket_name、data_folder三个列表的长度必须一致。"
   }
}

variable "bucket_name"{
    type = list(string)
    description = "Name or ARN of the bucket to put the file in"
}

variable "data_folder"{
    type = list(string)
    description = "Object path inside the bucket"
}

# 新增本地值构造展平的文件映射
locals {
  # 按索引配对三个配置项，统一处理路径末尾的斜杠
  upload_groups = [
    for i in range(length(var.source_file_path)) : {
      source_path = trimsuffix(var.source_file_path[i], "/")
      bucket      = var.bucket_name[i]
      target_path = trimsuffix(var.data_folder[i], "/")
    }
  ]
  # 遍历每个配置组，提取所有待上传文件构造唯一映射
  file_map = merge([
    for group in local.upload_groups : {
      # 第二个参数用**实现递归上传子目录，改为*则仅上传当前目录第一层文件
      for file in fileset(group.source_path, "**") : 
      "${group.bucket}/${group.target_path}/${file}" = {
        bucket      = group.bucket
        source      = "${group.source_path}/${file}"
        target_key  = "${group.target_path}/${file}"
      }
    }
  ]...)
}

# 调整后的上传资源块
resource "aws_s3_bucket_object" "upload_object"{
  for_each = local.file_map
  bucket   = each.value.bucket
  key      = each.value.target_key
  source   = each.value.source
  etag     = filemd5(each.value.source) # 基于文件内容生成校验值，仅文件变更时触发重传
}

你原有vars.tfvars的配置无需修改，直接使用即可实现需求。

内容的提问来源于stack exchange，提问作者Bien