Amazon EC2

Boto 2.x contains a number of customizations to make working with Amazon EC2 instances, storage and networks easy. Boto3 exposes these same objects through its resources interface in a unified and consistent way.

Creating the connection

Boto3 has both low-level clients and higher-level resources. For Amazon EC2, the higher-level resources are the most similar to Boto 2.x's ec2 and vpc modules:

# Boto 2.x
import boto
ec2_connection = boto.connect_ec2()
vpc_connection = boto.connect_vpc()

# Boto3
import boto3
ec2 = boto3.resource('ec2')

Launching new instances

Launching new instances requires an image ID and the number of instances to launch. It can also take several optional parameters, such as the instance type and security group:

# Boto 2.x

# Boto3
ec2.create_instances(ImageId='<ami-image-id>', MinCount=1, MaxCount=5)

Stopping and terminating instances

Stopping and terminating multiple instances given a list of instance IDs uses Boto3 collection filtering:

ids = ['instance-id-1', 'instance-id-2', ...]

# Boto 2.x

# Boto3

Checking what instances are running

Boto3 collections come in handy when listing all your running instances as well. Every collection exposes a filter method that allows you to pass additional parameters to the underlying service API operation. The EC2 instances collection takes a parameter called Filters which is a list of names and values, for example:

# Boto 2.x
reservations = ec2_connection.get_all_reservations(
    filters={'instance-state-name': 'running'})
for reservation in reservations:
    for instance in reservation.instances:
        print(instance.instance_id, instance.instance_type)

# Boto3
# Use the filter() method of the instances collection to retrieve
# all running EC2 instances.
instances = ec2.instances.filter(
    Filters=[{'Name': 'instance-state-name', 'Values': ['running']}])
for instance in instances:
    print(, instance.instance_type)

Checking health status of instances

It is possible to get scheduled maintenance information for your running instances. At the time of this writing Boto3 does not have a status resource, so you must drop down to the low-level client via ec2.meta.client:

# Boto 2.x
for status in ec2_connection.get_all_instance_statuses():

# Boto3
for status in ec2.meta.client.describe_instance_status()['InstanceStatuses']:

Working with EBS snapshots

Snapshots provide a way to create a copy of an EBS volume, as well as make new volumes from the snapshot which can be attached to an instance:

# Boto 2.x
snapshot = ec2_connection.create_snapshot('volume-id', 'Description')
volume = snapshot.create_volume('us-west-2')
ec2_connection.attach_volume(, 'instance-id', '/dev/sdy')

# Boto3
snapshot = ec2.create_snapshot(VolumeId='volume-id', Description='description')
volume = ec2.create_volume(, AvailabilityZone='us-west-2a')
ec2.Instance('instance-id').attach_volume(, Device='/dev/sdy')

Creating a VPC, subnet, and gateway

Creating VPC resources in Boto3 is very similar to Boto 2.x:

# Boto 2.x
vpc = vpc_connection.create_vpc('')
subnet = vpc_connection.create_subnet(, '')
gateway = vpc_connection.create_internet_gateway()

# Boto3
vpc = ec2.create_vpc(CidrBlock='')
subnet = vpc.create_subnet(CidrBlock='')
gateway = ec2.create_internet_gateway()

Attaching and detaching an elastic IP and gateway

Elastic IPs and gateways provide a way for instances inside of a VPC to communicate with the outside world:

# Boto 2.x

from boto.ec2.address import Address
address = Address()
address.allocation_id = 'eipalloc-35cf685d'

# Boto3

address = ec2.VpcAddress('eipalloc-35cf685d')