使用 Cloudformation 将 RDS 快照导出到 s3

Question

我想使用 cloudformation 将 RDS 快照导出到 s3，这将触发 lambda 函数。但我对此有点陌生。有人可以帮帮我吗。我已经为 Lambda 和 SNS 编写了脚本，但我很困惑如何使用 cloudformation 将 RDS 快照导出到 S3。我附上了从 SNS 触发 Lambda 的文件。帮助将不胜感激。谢谢。

AWSTemplateFormatVersion: '2010-09-09'
Transform: 'AWS::Serverless-2016-10-31'
Description: >
  example-lambda-sns
  Example CloudFormation template to subscribe a lambda to an SNS Topic.
Resources:
  ExampleTopic:
    Type: AWS::SNS::Topic
    Properties:
      DisplayName: !Sub 'example-sns-topic'
      TopicName: !Sub 'example-sns-topic'
      Subscription:
        - Protocol: lambda
          Endpoint: !GetAtt ExampleFunction.Arn
  
  ExampleFunction:
    Type: AWS::Serverless::Function
    Properties:
      FunctionName: mylambdafunction
      Description: mylambdafunction
      Handler: index.handler
      Runtime: nodejs14.x
      Role: arn:aws:iam::48
      CodeUri: 's3://testing/tes.zip'

  ExampleFunctionInvokePermission:
    Type: 'AWS::Lambda::Permission'
    Properties:
      Action: 'lambda:InvokeFunction'
      FunctionName: !Ref ExampleFunction
      Principal: sns.amazonaws.com   

  ExampleTopicPolicy:
    Type: 'AWS::SNS::TopicPolicy'
    Properties:
      Topics:
        - !Ref ExampleTopic
      PolicyDocument:
        Version: '2012-10-17'
        Statement:
          - Effect: Allow
            Action: 'sns:Publish'
            Resource: !Ref ExampleTopic
            Principal:
              AWS: '*'
            Condition:
              ArnLike:
                AWS:SourceArn: !Sub 'arn:aws:*:*:${AWS::AccountId}:*'```

Answer 1

以下是 lambda 函数的示例。这将从sns获得触发。创建快照时，SNS 会收到通知。

import json
import boto3
import uuid
import datetime
import os
def lambda_handler(event, context):
    snapshotarn = event['Records'][0]['Sns']['Message']
    snapshotarn=snapshotarn.split()
    snapshotarn=snapshotarn[10].replace('.','')
    print(snapshotarn)
    rds=boto3.client('rds')
    export=rds.start_export_task(
        ExportTaskIdentifier='export'+'-'+uuid.uuid4().hex,
        SourceArn=snapshotarn,
        S3BucketName=os.environ.get('S3_BUCKET_NAME'),
        IamRoleArn=os.environ.get('IAM_ROLE_FOR_EXPORT_TASK'),
        KmsKeyId=os.environ.get('KMS_KEY_ID'),
        S3Prefix=os.environ.get('S3_PREFIX'),
    )
    status={
        'ExportTaskIdentifier':export['ExportTaskIdentifier'],
        'S3Bucket':export['S3Bucket'],
        'S3Prefix':export['S3Prefix'],
        'Status':export['Status'],
        'ResponseMetadata':export['ResponseMetadata'], 
    }
    print(status)
    # TODO implement
    return {
        'statusCode': 200,
        'body': status
    }