jpbarto

AWS Tooling Catalog

The following is a living catalog of tools that I have found for managing AWS resources.

AWS Account Management

• AWS-Daleks - A tool for purging everything (almost) in an AWS account
• Cloud Custodian - Account governance tool built by Capital One
• AWS-Tailor - Collection of Lambda functions for configuring new AWS accounts

AWS Security

jpbarto

AWSTemplateFormatVersion: '2010-09-09'
Description: >
  A sample ElasticBeanstalk cloudformation template that deploys a 2 (MinSize) cluster of PHP servers 
  behind an application load balancer listening on both port 80 and 8080.  The ALB is placed (ideally)
  into a set of public subnets with the PHP servers in private subnets.

Parameters:
  ec2Subnets:
    Type: List<AWS::EC2::Subnet::Id>
    Description: Subnets for PHP servers

jpbarto

#!/usr/bin/python

from datetime import datetime, timedelta
import boto3

# region to query and what AWS credentials profile to use
REGION = 'eu-west-2'
PROFILE = 'default'

MONTH_PERIOD = 3600 * 24 * 30                               # ask CloudWatch to average over a 30 day

jpbarto

the following estimates the theoretical network latency for travelling roughly half way around the earth

circumference of earth = 40075 km

speed of light = 299792.458 km / s

( circumference of earth ) / 2 ) / ( speed of light ) = seconds to travel from one side of globe to another

(40075 / 2) / (299792.458) = 0.0668 seconds

67 ms ideal travel time from one side of globe to another

jpbarto

#!/usr/bin/env python3
"""
The following script demonstrates how to use the AWS Boto3 SDK to iterate through 
all of the users in an AWS Cognito User Pool and examine the events associated
with each user.  

If any failed authentication events are found the script formats them as messages
and logs them to CloudWatch logs.

This script could easily be modified to run periodically as a Lambda function 

jpbarto

##
## AMI created using 
## curl -sL https://rpm.nodesource.com/setup_11.x | bash -
## yum install -y git nodejs
## git clone https://github.com/socketio/socket.io.git
##
## cd socket.io
## npm install
## cd examples/chat
## npm install

jpbarto

jpbarto

import pandas as pd
import numpy as np
import argparse
from sklearn import preprocessing
import pickle
import os
from io import StringIO
from sagemaker_containers.beta.framework import (
    content_types, encoders, env, modules, transformer, worker)

jpbarto

{
    "TrainingJobName": "cli-t-job",
    "HyperParameters": {
        "sagemaker_submit_directory": "s3://sagemaker-eu-west-2-776347453069/sagemaker-scikit-learn-2019-06-19-13-21-02-471/source/sourcedir.tar.gz",
        "sagemaker_program": "sklearn_abalone_featurizer.py",
        "sagemaker_enable_cloudwatch_metrics": "false",
        "sagemaker_container_log_level": "20",
        "sagemaker_job_name": "sagemaker-scikit-learn-2019-06-19-13-21-02-471",
        "sagemaker_region": "eu-west-2"
    },

jpbarto

###
#
# CloudFormation Template to create EKS Worker Nodes without access to an IGW
#
# The following CloudFormation template is based upon the AWS-provided CloudFormation template for creating 
# EKS worker nodes in your VPC.  It differs in that it has user-data modified to configure a node to operate 
# within a VPC that has no IGW.  
#
# To use this template you will need a VPC with an EKS cluster already deployed and private access enabled. 
# You will also need a web proxy configured to allow the worker nodes access to the AWS EKS service API.  This