arbabnazar

boto3 quick hands-on

This documentation aims at being a quick-straight-to-the-point-hands-on AWS resources manipulation with [boto3][0].

First of all, you'll need to install [boto3][0]. Installing it along with [awscli][1] is probably a good idea as

• [awscli][1] is boto-based
• [awscli][1] usage is really close to boto's

arbabnazar

- name: Tell the host about our servers it might want to ssh to
  known_hosts:
    path: '/etc/ssh/ssh_known_hosts'
    name: 'github.com'
    key: "{{ lookup('pipe', 'ssh-keyscan -t rsa github.com') }}"

arbabnazar

# Extracting all ips from defined group
ips: "{{ groups['webservers']|map('extract', hostvars, ['ansible_eth1', 'ipv4', 'address'])|list }}"

# Iterate through dict, reject empty values
nsq_nsqd_opts: "{% for key, value in nsq_nsqd.iteritems() if value is not none %}-{{ key | replace('_', '-') }}=\"{{ value }}\" {% endfor %}"

arbabnazar

RDS Production Termination Protection

Create a policy like this, substituting your AWS account number, then attach it to all groups and roles. It will prevent deletion of RDS instances containing the string "prod" in their name.

Note: This is intended to prevent accidental deletion, and is easily sidestepped.

arbabnazar

#!/usr/bin/expect

set timeout -1;
spawn {{django_dir}}/venv/bin/python manage.py changepassword {{admin_user}};
expect {
    "Password:" { exp_send "{{admin_pass}}\r" ; exp_continue }
    "Password (again):" { exp_send "{{admin_pass}}\r" ; exp_continue }
    eof
}

arbabnazar

Session Management in an Autoscaling Environment

Problem Statement

User sessions in J2EE and LAMP stacks have traditionally been handled in memory by the application server handling the user request. Because of that, load balancers have been configured to use sticky sessions. By sticky sessions we mean that once the user has visited the site, they will be assigned an app server and will return to that server for subsequent requests. The load balancers typically handle that by referencing the users session cookie.

Elastic cloud environments differ from traditional server configurations in that they have a variable number of servers based on traffic loads whereas traditional configurations had a fixed number of servers. When traffic volumes decline it is necessary to vaporize servers. In doing so, we would lose user sessions (essentially forcing a logout) unless we come up with a new strategy for session management.

A new approach

After much research, it is clear that the best

arbabnazar

---
- name: Ansible tips
  connection: local
  hosts: localhost

  tasks:
    - name: Ping only if OSX
      ping:
      when: "{{ lookup('pipe', 'uname -s') | match('Darwin') }}"

arbabnazar

#!/bin/bash

apt-get update
apt-get install -y libffi-dev g++ libssl-dev python-pip python-dev git
pip install -U awscli ansible setuptools


aws s3 cp s3://tendo-github-key-s3/git-private-key /root/.ssh/bitbucket_secret_key
chmod 400 /root/.ssh/bitbucket_secret_key

arbabnazar

import re
import boto3
import csv
from botocore.exceptions import ClientError

ec2 = boto3.client('ec2')

def get_snapshots():
    return ec2.describe_snapshots(OwnerIds=['self'])['Snapshots']

arbabnazar

import jenkins.model.*
import hudson.model.*
import jenkins.plugins.nodejs.tools.*
import hudson.tools.*

def inst = Jenkins.getInstance()

def desc = inst.getDescriptor("jenkins.plugins.nodejs.tools.NodeJSInstallation")

def versions = [