When working with the AWS EC2 API (particularly describe-instances), you'll encounter reservation IDs before seeing actual instance IDs. This architectural design reflects how EC2 internally manages instance launches:

# Sample describe-instances response structure
{
    "Reservations": [
        {
            "ReservationId": "r-1234567890abcdef0",
            "Instances": [
                {
                    "InstanceId": "i-1234567890abcdef0",
                    # ... other instance details
                }
            ]
        }
    ]
}

The reservation ID becomes particularly important in these scenarios:

• Bulk launches: When using RunInstances with MaxCount > 1
• Auto Scaling Groups: Instances launched in the same scaling event
• Spot Fleets: Multiple spot instances requested simultaneously

# Launching multiple instances in one request
aws ec2 run-instances \
    --image-id ami-0abcdef1234567890 \
    --count 3 \
    --instance-type t3.medium
# All 3 instances will share the same reservation ID

For robust instance filtering in your snapshot automation:

import boto3

def get_running_instances():
    ec2 = boto3.client('ec2')
    response = ec2.describe_instances(Filters=[{'Name': 'instance-state-name', 'Values': ['running']}])
    
    instances = []
    for reservation in response['Reservations']:
        for instance in reservation['Instances']:
            instances.append(instance['InstanceId'])
    
    return instances

Key differences:

• Terminated instances may temporarily appear in reservations
• Spot instances can have complex reservation patterns
• Cross-account sharing affects reservation visibility

When working with AWS EC2's describe-instances API call, you'll notice the response structure groups instances under "Reservations". This isn't about Reserved Instances (the billing concept) - it's an implementation detail of EC2's launch architecture.

# Sample describe-instances response structure
{
    "Reservations": [
        {
            "ReservationId": "r-1234567890abcdef0",
            "Instances": [
                {
                    "InstanceId": "i-1234567890abcdef0",
                    # ... other instance details
                }
            ]
        }
    ]
}

A single reservation ID can represent multiple instances in these scenarios:

• Bulk launches: When using RunInstances with MaxCount > 1
• Auto Scaling Groups: Instances launched together in a scaling event
• Spot Fleets: Multiple spot instances requested simultaneously

# Launching multiple instances in one request
aws ec2 run-instances \
    --image-id ami-0abcdef1234567890 \
    --count 3 \
    --instance-type t3.medium
# These 3 instances will share one ReservationId

When processing describe-instances output, you need to:

1. Iterate through the Reservations array first
2. Then process each Instance within the reservation

# Python example processing describe-instances response
import boto3

ec2 = boto3.client('ec2')

def get_running_instances():
    response = ec2.describe_instances(Filters=[{'Name': 'instance-state-name', 'Values': ['running']}])
    
    instance_ids = []
    for reservation in response['Reservations']:
        for instance in reservation['Instances']:
            instance_ids.append(instance['InstanceId'])
    
    return instance_ids

Key behaviors to understand:

• A reservation ID persists through instance stops/starts
• Terminating all instances in a reservation doesn't immediately clear the reservation from API responses
• Reservation IDs are region-specific

The reservation ID essentially represents the initial launch context of your instances, which is why it appears in monitoring and API responses even though it's not a concept you typically interact with directly in most workflows.