如何通过AWS Lambda实现端口转发以访问本地数据库？

Absolutely! You can totally use AWS Lambda to access your local MySQL database for data analysis—here's how to pull it off, plus key considerations to make the setup smooth.

Yes, Lambda absolutely can connect to databases outside AWS (including your local MySQL instance). Since Lambda runs in AWS's managed environment, you just need to bridge the network gap between Lambda and your local machine, similar to how you did with EC2. Here are the most practical approaches:

Option 1: Replicate Your EC2 Port Forwarding Setup (Quickest for Testing) #

This mirrors what you already did with EC2, using a proxy instance to route Lambda traffic to your local DB:

• First, make sure your local machine is reachable from AWS: You’ll need a public IP (or use a VPN to link your local network to AWS) and open port 3306 in your local firewall to allow traffic from the proxy.
• Deploy an EC2 instance in your AWS VPC to act as the proxy. This instance will forward incoming traffic from Lambda to your local DB.
• On the EC2 proxy, set up port forwarding with a tool like socat or SSH. For example, run this command to forward traffic from the proxy’s port 3306 to your local DB:

  socat TCP-LISTEN:3306,fork TCP:your-local-public-ip:3306
  

• Attach your Lambda function to the same VPC as the EC2 proxy, and assign it a security group that allows outbound traffic to the proxy’s port 3306.
• In your Lambda Python code, connect to the EC2 proxy’s private IP:3306 exactly like you connected to your local DB from EC2.

Option 2: Use a VPN/Direct Connect (More Secure for Regular Use) #

If you plan to run this analysis frequently, a more stable and secure approach is to link your local network directly to AWS:

• Set up an AWS Site-to-Site VPN or AWS Direct Connect between your local network and your AWS VPC. This makes your local database act like it’s part of the AWS network.
• Attach your Lambda function to the VPC, and update your local firewall to allow inbound MySQL traffic from the AWS VPC’s CIDR range.
• Now your Lambda can connect directly to your local DB’s private IP (no proxy needed) just like it would connect to an AWS RDS instance.

Critical Lambda Tips to Remember #

• VPC Attachment is Mandatory: Lambda functions not attached to a VPC can’t reach external resources like your local DB. You must attach your function to a VPC with access to your local network.
• Cold Starts: Lambda cold starts take longer when attached to a VPC (since it needs to provision network interfaces). If your analysis runs often, use Provisioned Concurrency to cut down on wait times.
• Secure Credentials: Never hardcode your MySQL username/password in Lambda code. Store them in AWS Secrets Manager, and grant your Lambda execution role permission to retrieve the secrets.
• Latency: Since Lambda runs in AWS, there will be some network latency compared to running code locally. Test a few queries to make sure performance meets your needs.

Example Lambda Python Code Snippet #

Here’s a quick example of how your function might connect to the proxy (or local DB via VPN):

import pymysql
import boto3
import json

def lambda_handler(event, context):
    # Pull DB credentials from Secrets Manager
    secrets_client = boto3.client('secretsmanager')
    secret_response = secrets_client.get_secret_value(SecretId='your-mysql-secret-name')
    secret_details = json.loads(secret_response['SecretString'])
    
    # Connect to the database (use proxy IP or local DB IP via VPN)
    db_conn = pymysql.connect(
        host='proxy-ec2-private-ip',
        user=secret_details['username'],
        password=secret_details['password'],
        database='your-database-name',
        port=3306
    )
    
    # Run your analysis query
    with db_conn.cursor() as cursor:
        cursor.execute("SELECT * FROM your_target_table LIMIT 20")
        query_results = cursor.fetchall()
    
    db_conn.close()
    return {'statusCode': 200, 'analysis_results': query_results}

内容的提问来源于stack exchange，提问作者Nole