如何用Boto3将运行中EC2实例信息转换为Pandas DataFrame表格

Let's get your code working smoothly—there are a couple of small tweaks needed, like fixing the iterator exhaustion issue and implementing the missing get_name_tag method. Here's the full, functional version:

import boto3
import pandas as pd

# Initialize EC2 resource (or client, depending on your setup)
ec2 = boto3.resource('ec2')

def get_name_tag(instance):
    """Helper function to extract the 'Name' tag from an EC2 instance"""
    for tag in instance.tags or []:
        if tag['Key'] == 'Name':
            return tag['Value']
    # Return a default if no Name tag exists
    return 'Unnamed Instance'

# Get all running instances, store as a list to avoid iterator exhaustion
running_instances = list(ec2.instances.filter(
    Filters=[{'Name': 'instance-state-name', 'Values': ['running']}]
))

# Count instances (now we can use the list length)
instance_count = len(running_instances)
print(f"Total running EC2 instances: {instance_count}")

# Build our list of instance data
instance_data = []
for instance in running_instances:
    instance_info = {
        "name": get_name_tag(instance),
        "id": instance.id,
        "type": instance.instance_type,
        "launch_time": instance.launch_time.strftime('%Y-%m-%d %H:%M:%S'),  # Add extra useful field
        "availability_zone": instance.placement['AvailabilityZone']
    }
    instance_data.append(instance_info)

# Convert to Pandas DataFrame
ec2_df = pd.DataFrame(instance_data)

# Optional: Display the DataFrame (for Jupyter/console)
print(ec2_df)

Key Fixes & Improvements: #

• Iterator Exhaustion Fix: Originally, you called instances.all() twice—once for counting, once for looping. Since instances is an iterator, it gets emptied after the first use. By converting it to a list upfront (list(ec2.instances.filter(...))), we can reuse the data safely.
• get_name_tag Implementation: Added the missing helper function to pull the instance's Name tag (returns a default if no tag exists).
• Extra Useful Fields: Added launch_time and availability_zone to make the table more informative—feel free to add/remove fields based on your needs (like private IP, security groups, etc.).
• Cleaner Structure: Renamed variables to follow Python's snake_case convention for readability.

Once you run this, you'll get a nicely structured DataFrame with all your running EC2 instance details, ready for analysis, exporting to CSV, or further processing.

内容的提问来源于stack exchange，提问作者Ken J