Bottom Line: When I get a new Raspberry Pi, here is how I get it up and running with SSH over an ethernet cable directly attached to my MacBook.

I’ve really had a lot of fun over the last year or two learning some of the basics of Linux and the command line with the Raspberry Pi. The process has definitely made me a more capable Mac user, and I now have a Chromebook running Arch Linux as well.

In short, the Raspberry Pi Foundation is a non-profit organization whose goal was to develop a computer that would be affordable to people worldwide and facilitate learning computer science, programming, and electronics. Their two most recent models are the B+, which is credit-card sized and runs $35, and the slightly smaller A+, which is even more energy efficient and runs $25. Because they are UK based, shipping adds a bit to the cost from their distributors, but currently you can buy a B+ through Amazon for ~$33 with free shipping (not Prime) and an A+ for ~$25 with Prime shipping.

It has a great HDMI output and has a powerful enough graphics chip to display 1080p. That said, I’ve never hooked mine up to a monitor. Instead, all I need to get up and running is:

  • a micro USB cable for power
  • a micro SD card for storage
    • I like this $11, prime-eligible 16GB card from Amazon
    • If you’re using a RPi B instead of a B+ you’ll need a different kind of card, it doesn’t use the micro SD cards
    • The micro SD card I linked to also has a full-size SD card adapter, which is necessary to program it from the MBP SD card slot. If your computer doesn’t have an SD card slot and / or your micro SD card doesn’t come with an adapter, you’ll also need to pick up some kind of USB-to-micro-SD device.
  • an ethernet cable

Basically, aside from the bare minimum (power and SD card), the ethernet cable is sufficient to program everything over SSH from the MacBook. NB: My technique will need adjustment for folks with newer MacBooks (Airs and such) that don’t have an ethernet port!

The steps below should be used as a general guide and not necessarily followed verbatim. Please read my disclaimer before you blindly follow any of these steps! This is not intended to be run all at once as a shell script, instead each step should be run individually, especially since some are intended to be run from your Mac and others from the RPi. Most of the code blocks ( looks like this ) are intended to be run from on your Mac.

Setting up the SD card

From Mac, with micro SD card connected

  1. Torrent latest Raspbian distro:
  2. Unzip / extract the image
  3. Insert SD card into Mac
  4. Identify the /dev/disk number of the SD card: diskutil list
  5. Replace # with the appropriate number diskutil unmountDisk /dev/disk#
  6. Double check that number, since the next step could erase your MacBook’s hard drive if you got it wrong
  7. Copy the Raspbian image to the SD card: sudo dd if=raspbian.img of=/dev/rdisk# bs=1m && sync
  8. diskutil eject /dev/disk#
  9. Remove the SD card and insert into the RPi

Setting up the RPi, still from the OS X Terminal

After connecting RPi directly with Ethernet cable but before powering it on:

  1. On Mac turn on System Preferences -> Sharing -> Internet Sharing (make sure ethernet is checked)
  2. Turn on the RPi
  3. Output the RPi’s IP address with arp -a | grep bridge | grep -v -e incomplete -e 255 | grep -oE '\d+\.\d+\.\d+\.\d+'
    1. If the above command doesn’t work, you may have to break it down into manual steps by:
    2. Connect everything and power up Pi
    3. Turn off Internet Sharing and wait a few seconds
    4. arp -a
    5. Turn on Internet Sharing, wait about a minute
    6. arp -a
    7. Comparing the output from the two arp -as, look for a new address that probably involves the word “bridge”, likely one without 255 anywhere, and there’s a good chance it starts with 192.168
  4. SSH into the address you found using username pi, i.e. ssh [email protected]
  5. You should probably write down that IP address as well
  6. Default password is raspberry
  7. sudo raspi-config
  8. Go through every option. My notes:
    • First thing first: Expand Filesystem
    • Locale: Deselect the GB one, select en_US.UTF-8 UTF-8
    • Advanced -> hostname, add a friendly (simple character only) name you want to call your Pi, e.g. fred or mypi
    • Consider changing GPU memory — mine run headless, so I distribute all the memory away from the GPU
  9. Reboot sudo reboot and reconnect by ssh
  10. Update the available packages: sudo apt-get update
  11. Upgrade available updates: sudo apt-get upgrade
  12. Upgrade the RPi firmware: sudo rpi-update
  13. sudo reboot

Start customizing your installation

Install important packages and tools

  1. Become superuser (instead of the pi user): sudo su
  2. Install your basic set of important utilities and packages (yours should be different, as you learn what tools you like to use): aptitude install vim git monit avahi-daemon libssl-dev i2c-tools watchdog python3
  3. Install ack (very optional, can also just aptitude install ack-grep, but I prefer this way)
    • curl -L | perl - App::cpanminus
    • cpanm App::Ack

Make a new user (so you can eventually delete pi — important for security), still as superuser / root

  1. Replace USER with your username of choice: adduser USER
  2. Set a new password for your new user: passwd USER
  3. Set an editor for use with visudo: update-alternatives --set editor /usr/bin/vim.tiny
    • Warning: vim is my editor of choice but is pretty tough to use and certainly not intuitive. If you’re unfamiliar with vim, Google some alternative text editors, install them (or verify they are installed), and replace /user/bin/vim.tiny with something like --set editor $(which YOUR_EDITOR)
  4. Edit the (very important) sudoers file to set who can have superuser (e.g. su, sudo) privileges: visudo
    • Copy root ALL=(ALL:ALL) ALL to a new line
    • In that copied line, change root to your new USER
    • Save and exit (:wq in vim)

Optional: If you intend to use the GPIO, still as superuser

  1. vim /etc/modules
    • add i2c-bcm2708
    • add i2c-dev
    • add bcm2708_wdog
  2. vim /etc/modprobe.d/raspi-blacklist.conf and comment out:
    • blacklist spi-bcm2708
    • blacklist i2c-bcm2708
    • Save and exit

Optional: Set up watchdog (to automatically reboot the Pi if it freezes — very rare, but just in case), still as superuser

  1. update-rc.d watchdog defaults
  2. vim /etc/watchdog.conf
    • Uncomment watchdog-device = /dev/watchdog
    • Uncomment max-load-1 = 24 (e.g. it would take 24 RPis to complete a given in 1 minute)
    • Save and exit

Very optional: install python3.4 from source (using my Gist), still as superuser

  1. wget
  2. bash
  3. Install important packages: /opt/python3.4.2/bin/pip3 install RPi.GPIO pyserial wiringpi2

Verify your new user and password is working

  1. reboot
  2. From your Mac, using your new USER and password: ssh [email protected]
  3. If it works, this command should both verify that you have properly given yourself sudo privileges and also delete the pi user and home directory: sudo userdel -r pi
  4. If that worked, you can sudo visudo and delete the line pi ALL=(ALL) NOPASSWD: ALL (and any other line involving the pi user)

At this point you’re up and running! My next step is often to connect up a WiFi adapter and set it up over the ethernet cable, after which you can find the new WiFi card’s IP address, lose the ethernet cable, turn off internet sharing on your Mac, and continue to access the Pi via SSH from WiFi. Have fun!