Andrew Odewahn odewahn

How do we make it easy (well, easier) for users to run pre-built Docker containers?

Binder

Launchbot

Desktop GUI for publishing and running containers.

Install this:

https://github.com/rackerlabs/carina/releases

Copy it to a directory on your path and make it executable

$ mv carina-darwin-amd64 /usr/local/bin/carina

Network Notes

Set up a Docker of Vagrant-based lab where you set up your own network and topologies to learn, with a final case study on setting up a secure (!) VPC on AWS.

tcp/ip on Linux

IP addresses
sub-nets and CIDR
IPv6
Ports

	package main

	import (
	"crypto/rand"
	"fmt"
	"log"

	"github.com/fsouza/go-dockerclient"
	)

	package main

	import (
	"fmt"
	"os/exec"
	"strings"
	)

	func runCmd(cmdArgs []string) ([]string, error) {
	cmdName := "docker-machine"

	export TOKEN=$( head -c 30 /dev/urandom \| xxd -p )
	export POOL_SIZE=5
	export OVERPROVISION_FACTOR=2
	export CPU_SHARES=$(( (1024*${OVERPROVISION_FACTOR})/${POOL_SIZE} ))
	export TMPNB_NODE=6df651e0-4701-411d-b11c-3cbfe73b86d7-n1
	export DOCKER_HOST=tcp://104.130.0.38:2376

	docker run -d \
	-P \
	-h spark.tmpnb-oreilly.com \

	export TOKEN=$( head -c 30 /dev/urandom \| xxd -p )
	export POOL_SIZE=5
	export OVERPROVISION_FACTOR=2
	export CPU_SHARES=$(( (1024*${OVERPROVISION_FACTOR})/${POOL_SIZE} ))
	export TMPNB_NODE=bb0f0ccc-2dcb-410b-94cf-808c99324ab6-n1
	export DOCKER_HOST=tcp://104.130.0.25:2376

	docker run -d \
	-P \
	-p 80 \



	# Set up a working ipython environment

	The simplest way to do this is to install Anaconda:

	http://docs.continuum.io/anaconda/install.html

	# Install IPython3

	<!DOCTYPE html>
	<html lang="en">
	<head>
	<meta charset="utf-8">
	<title>D3: Data-driven circles</title>
	<script type="text/javascript" src="d3.v3.js"></script>
	<style type="text/css">

	body {
	background-color: gray;