Windows下配置boto3/boto库及Python操作AWS EC2实例问题咨询

Hey there! Let's break down your AWS Python questions one by one, since you're already set up with Python 3.6.4 on Windows 8.1 64-bit.

1. Installing and Configuring boto3 & boto Libraries #

First, let's get the libraries installed, then set up your AWS credentials so the SDK can authenticate with your account.

Installation Steps #

Open Command Prompt (or PowerShell) and run this command to install both libraries. Using py -3.6 ensures you target your specific Python 3.6 installation if you have multiple versions on your system:

py -3.6 -m pip install boto3 boto

If you only have Python 3.6 installed, a simpler pip install boto3 boto will work too.

Configuration Options #

You have two main ways to set up AWS credentials for boto3/boto:

Option 1: Local Credentials Files

Create a .aws folder in your Windows user directory (e.g., C:\Users\YourUsername\.aws). Inside this folder, make two files (no extensions):

• credentials:

  [default]
  aws_access_key_id = YOUR_AWS_ACCESS_KEY
  aws_secret_access_key = YOUR_AWS_SECRET_KEY
  

• config:

  [default]
  region = us-east-1  # Replace with your preferred default region (e.g., eu-west-1)
  

Make sure to replace the placeholders with your actual AWS credentials, which you can grab from the AWS IAM console.

Option 2: Environment Variables

If you prefer not to use files, set these temporary environment variables in Command Prompt:

set AWS_ACCESS_KEY_ID=YOUR_ACCESS_KEY
set AWS_SECRET_ACCESS_KEY=YOUR_SECRET_KEY
set AWS_DEFAULT_REGION=us-east-1

For permanent settings, go to Control Panel > System > Advanced System Settings > Environment Variables and add them there.

2. Starting All AWS EC2 Instances with boto3 #

Here's a script to start all stopped EC2 instances (we filter for stopped ones to avoid errors from trying to start already running instances):

import boto3
from botocore.exceptions import ClientError

def start_all_stopped_instances(region_name=None):
    # Initialize EC2 client; use default region if none specified
    ec2 = boto3.client('ec2', region_name=region_name)
    
    try:
        # Fetch all instances
        response = ec2.describe_instances()
        stopped_instance_ids = []
        
        # Extract IDs of stopped instances
        for reservation in response['Reservations']:
            for instance in reservation['Instances']:
                if instance['State']['Name'] == 'stopped':
                    stopped_instance_ids.append(instance['InstanceId'])
        
        # Start instances if any are stopped
        if stopped_instance_ids:
            ec2.start_instances(InstanceIds=stopped_instance_ids)
            print(f"Successfully started {len(stopped_instance_ids)} instances: {', '.join(stopped_instance_ids)}")
        else:
            print("No stopped instances found to start.")
    except ClientError as e:
        print(f"Error starting instances: {e.response['Error']['Message']}")

# Example usage: start instances in default region
start_all_stopped_instances()
# Or specify a region: start_all_stopped_instances(region_name='eu-west-1')

If you want to start all instances regardless of state (even though this will throw errors for running ones), just remove the state filter and collect all instance IDs.

3. Fixing Issues with Retrieving & Stopping EC2 Instances in a Specific Region #

Let's troubleshoot common causes step by step:

Step 1: Verify Your Region Configuration #

Double-check you're targeting the correct region. If you don't specify it in code, boto3 uses the default from your config file or environment variables. To explicitly set it:

ec2 = boto3.client('ec2', region_name='your-target-region')  # e.g., 'ap-southeast-1'

AWS region names are case-sensitive (e.g., us-east-1 not USEAST1), so make sure the spelling is exact.

Step 2: Check IAM Permissions #

Your AWS user needs permissions to describe and stop instances. Create or attach an IAM policy with these permissions:

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "ec2:DescribeInstances",
                "ec2:StopInstances"
            ],
            "Resource": "*"
        }
    ]
}

You can attach this policy via the AWS IAM console to your user or role.

Step 3: Debug with Error Handling #

Add error handling to your script to pinpoint the issue (permission error, missing instances, etc.):

import boto3
from botocore.exceptions import ClientError

def stop_instances_in_region(region_name):
    ec2 = boto3.client('ec2', region_name=region_name)
    
    try:
        # Fetch all instances in the region
        response = ec2.describe_instances()
        instance_ids = [instance['InstanceId'] for reservation in response['Reservations'] for instance in reservation['Instances']]
        
        if not instance_ids:
            print(f"No instances found in region {region_name}.")
            return
        
        # Stop instances
        ec2.stop_instances(InstanceIds=instance_ids)
        print(f"Stopped {len(instance_ids)} instances in {region_name}: {', '.join(instance_ids)}")
    except ClientError as e:
        print(f"Error: {e.response['Error']['Message']}")
        print(f"Error Code: {e.response['Error']['Code']}")

# Test with your target region
stop_instances_in_region('your-target-region')

Common error codes to watch for:

• AccessDenied: Your user lacks permissions
• InvalidRegionID: Incorrect region name
• InvalidInstanceID.NotFound: Targeted instances don't exist in that region

Step 4: Check Network/Proxy Settings #

If you're on a corporate network, a proxy might block traffic to AWS APIs. Configure a proxy in boto3 like this:

import boto3
from botocore.config import Config

proxy_config = Config(
    proxies={
        'http': 'http://your-proxy-address:port',
        'https': 'https://your-proxy-address:port'
    }
)

ec2 = boto3.client('ec2', region_name='your-region', config=proxy_config)

Alternatively, set the HTTP_PROXY and HTTPS_PROXY environment variables in Windows.

内容的提问来源于stack exchange，提问作者zlogic