boto3 Cheat Sheet (DRAFT) by wtranmer

To use boto3, you must first install python. There are a number of distri­butions available; among the most popular are Anaconda3, Active­Python, and PyPy.

I prefer Anaconda3 because of the large number of pre-in­stalled packages. This does however also mean it is one of the largest distri­but­ions. But as the saying goes, "It is better to have and not need than to need and not have."

Once you have installed your Python of choice, use the pip installer to install boto3.

pip install boto3

Once boto3 is installed, install the Amazon AWS CLI tools and run aws configure to set your creden­tials and default region.

aws configure

Now you are ready to roll.

boto3 includes access to almost all of the AWS services. To interact with these services, you create a resource or client object that connects to a particular service. Then you can use the service using boto3's api for that object.

For instance, to create a new EC2 instance, you will create a resource object that is connected to 'ec2.' Then with that object, you will call a function to create the instance and pass the approp­riate parameters to the function.

Connect to EC2
To create an EC2 resource object

import boto3

ec2 = boto3.r­es­our­ce(­'ec2')

instances = ec2.in­sta­nce­s.all()

# Print raw list of instances

for instance in instances:

 ­ ­print( instance )

# Print list of instances by name tag

for instance in instances:

 ­ ­print( [tag['­Value'] for tag in instan­ce.tags if tag.ge­t('­Key') == 'Name'] )

# Print list of instances by state

states = (

 ­ { 'Code' : 0, 'state' : 'pending' },

 ­ { 'Code' : 16, 'state' : 'running' },

 ­ { 'Code' : 32, 'state' : 'shutt­ing­-down' },

 ­ { 'Code' : 48, 'state' : 'termi­nated' },

 ­ { 'Code' : 64, 'state' : 'stopping' },

 ­ { 'Code' : 80, 'state' : 'stopped' },

)

instan­ce_­states = {

 ­ ­'pe­nding': [],

 ­ ­'ru­nning': [],

 ­ ­'sh­utt­ing­­down' : [],

 ­ ­'te­rmi­nated' : [],

 ­ ­'st­opping' : [],

 ­ ­'st­opped' : [],

}

for instance in instances:

 ­ ­ins­tan­ce_­nam­e=[­tag­['V­alue'] for tag in instan­ce.tags if tag.ge­t('­Key') == 'Name']

 ­ ­ins­tan­ce_­sta­tes­[in­sta­nce.st­ate­['N­ame­']].ap­pend( instan­ce_­name[0] )

for i in instan­ce_­states:

 ­ ­ins­tan­ce_­sta­tes­[i].sort()

 ­ ­print(i + ' instan­ces­\n--')

 ­ for name in instan­ce_­sta­tes[i]: print( name )

 ­ ­pri­nt('')

# Gather instance details

# ID of the AMI (Amazon Machine Image) that the VM will use for OS

`# This is the ami for Window Server 2012 R2 64-bit

ami_id = ami-3d­787d57

# IDs of the Security Groups to be assign to the VM

securi­ty_­gro­up_ids = []

# Single subnet id in which VM will be started

subnet_id = 

# Key file that will be used to decrypt the admini­strator password

keyfil­e_name = 

# ARN of the IAM instance profile to be assigned to the VM

iam_pr­ofi­le_arn = 'arn:a­ws:­iam­::1­234­567­890­12:­ins­tan­ce-­pro­fie­l/m­ypr­ofile

# VM instance type / instance size

instan­ce_type = 't2.micro'

# Disk drive capacity in GB

disk_size = 120

# UserData. Code block to execute on first instance launch

user_data = ' '

# Create the instance

ec2.cr­eat­e_i­nst­ances(

 ­ ­ImageId = ami_id,

 ­ ­Min­Count = 1,

 ­ ­Max­Count = 1,

 ­ ­KeyName = keyfil­e_name,

 ­ ­Sec­uri­tyG­roupIds = securi­ty_­gro­up_ids,

 ­ ­Ins­tan­ceType = instan­ce_­type,

 ­ ­Blo­ckD­evi­ceM­appings = [

 ­ ­ ­ {

 ­ ­ ­ ­'De­vic­eName': '/dev/­sda1',

 ­ ­ ­ ­'Ebs': {

 ­ ­ ­ ­ ­ ­'Vo­lum­eSize': disk_size,

 ­ ­ ­ ­ ­ ­'De­let­eOn­Ter­min­ation': True,

 ­ ­ ­ ­ ­ ­'Vo­lum­eType': 'gp2', 

 ­ ­ ­ }

 ­ ],

 ­ ­Iam­Ins­tan­ceP­rofile = {

 ­ ­ ­ ­'Arn': arn_pr­ofile

 ­ },

 ­ ­Sub­netId = subnet_id,

 ­ ­Use­rData = user_data

)