Working with Raspberry PI OS

Installing OS on SD Card

• Install Raspberry PI OS Imager from https://www.raspberrypi.com/software/.
• Plug the memory card with the computer.
  NOTE: Don't mistakenly plug the memory card with Raspberry PI and try to install the OS.
• Run the Imager program.
• From Imager settings, enable and enter hostname, SSH, Wifi, eventually configure everything.
  Note: if you already have another active PI device, then make sure hostname are different.

How to choose the correct version of OS

• If you use the OS headless (i.e. without monitor), then install the lite version. Otherwise Recommended version is good to go.
• If you want a bootable memory card to use in both 32bit and 64bit Raspberry pi devices, then you have to install 32bit version of OS.

How to connect and work with Raspberry PI from a computer from command line (i.e. headleass)

• To enable headless (i.e. command-line/SSH) access, after installing the OS, create an empty file ssh in the root directory of the memory card.
• Plug the memory card to Raspberry PI.
• Connect to the power either using a 5v charger or to the computer usb.
• Wait for a couple of minutes to boot the OS. It will automatically connect to the WIFI. Note: to change or add WIFI information, check Usefull Notes below.
• Go to the Router admin panel and get the IP address of the RPI to use as the hostname.
• Putty can be used to access the RPI from command line or VSCode has extensions for that. Install entire Remote Development suite VSCode extensions (It will install a set of extensions. By Microsoft) to access and to do development in Raspberry PI. Also the extension SSH-FS is a useful one.
• Start creating a new remote connection on the extension.
• When prompted for ssh user@hostname, enter in that format ssh -p 22 <os_username>@<rpi_ip_address>.

Update and upgrade Linux package information

First things first:

sudo apt-get update
sudo apt-get upgrade

sudo gives root privilege.

Install Git

sudo apt install git

Enter the command below to verify the installation:

git --version

Install Node.js

Install Node.js from the NodeSource Repository, a third party service which resolves the installation process.

curl -sL https://deb.nodesource.com/setup_<version>.x | sudo bash -

sudo apt-get install -y nodejs

Go to https://deb.nodesource.com/ for command with latest version.

NOTE

• Try avoiding the process of downloading installer, extracting etc. For new Linux users, it can be a mess.
• ARMv6 processor is not supported by NodeSource; even for Node-v10.

NodeJS Install Steps for ARMv61 processor (RPi Zero W)

1. Go to the unoffical builds download page of Node.js site and select the version of NodeJS you want to install. Copy the link of the ARMv6 version of NodeJS (Select the link with the .xz extension).

2. Download the file wget <link>. Example

wget https://unofficial-builds.nodejs.org/download/release/v14.13.0/node-v14.13.0-linux-armv6l.tar.xz

1. Extract the binary from the tarball file using the following command tar xvfJ <file_name.tar.xz>. Example

tar xvfJ node-v14.13.0-linux-armv6l.tar.xz

1. Copy the contents of the extracted tarball file to the usr/local directory: sudo cp -R <extracted tar folder>/* /usr/local. Example:

sudo cp -R node-v14.13.0-linux-armv6l/* /usr/local

1. Reboot and check everything is working correctly

   node -v && npm -v`.

Extra steps

1. You might have to do the following if you receive a command not found error. First, open the .profile file using nano: sudo nano ~/.profile
2. Add the following line to the end of the file and hit Ctrl+X to save, and then hit 'y' and enter to confirm the changes: PATH=$PATH:/usr/local/bin.
3. Reboot and check is it working. NOTE: If running node command shows exec format error then undo the previous steps and try again. Make sure to add PATH variable to .profile and then reboot.

Command to install Python package installer (pip3) and python packages

sudo apt install python3-pip

pip3 install smbus2

Execute certain command on boot

• Edit /etc/rc.local with root permission and commands before exit command.

Example

   node /home/pi/projects/raspberry-pi-projects/home-iot/app.js &

To know about the ending &, check How to run a process in the background in this doc.

Restart network or the OS when connection is lost

• Create a shell script /usr/local/bin/checkwifi.sh.

  ping -c4 <router_ipv4> > /dev/null
  
  if [ $? != 0 ]
  then
    echo "No network connection, restarting wlan0"
    /sbin/ifdown 'wlan0'
    sleep 5
    /sbin/ifup --force 'wlan0'
  fi

  • Open the crontab editor by typing:

  crontab -e

  Add the following line:

  */5 * * * * /usr/bin/sudo -H /usr/local/bin/checkwifi.sh >> /dev/null 2>&1

  This will run the script in checkwifi.sh every 5 minutes, writing its output to /dev/null so it won't clog your syslog.

  • To reboot the PI, write the following script instead in checkwifi.sh:

  ping -c4 <router_ipv4> > /dev/null
   
  if [ $? != 0 ]
  then
    sudo /sbin/shutdown -r now
  fi

Enable the 1-Wire Interface

(Courtesy: circuitbasics.com)
We’ll need to enable the One-Wire interface before the Pi can receive data from the sensor. Once you’ve connected the DS18B20, power up your Pi and log in, then follow these steps to enable the One-Wire interface:

• At the command prompt, enter

sudo nano /boot/firmware/config.txt

then add this text to the bottom of the file:

dtoverlay=w1-gpio

Note: If any dtoverlay= is already there, doesn't matter; add one more at bottom.

• Exit nano, and reboot the Pi with sudo reboot.

• Log in to the Pi again, and at the command prompt enter sudo modprobe w1-gpio. Then enter sudo modprobe w1-therm.

  • Modules are pieces of code which extend the functionality of the operating system kernel without the need to reboot. Once loaded, modules reside in memory.
  • To remove the module use sudo modprove -r <module_name>

• Change directories to the /sys/bus/w1/devices directory.

• Enter ls to list the devices. Something like 28-xxxxxxxxxxx w1_bus_master1 will be displayed.

• Go to 28-xxxxxxxxxxx directory and type cat w1_slave to see the temperature value from sensor.

[Extra] Install LocalTunnel (proxy tool)

1. Install localtunnel globally to use localtunnel command lt from anywhere

npm install -g localtunnel

If not installed globally then full path to lt executable have to be used, which will be something like /home/pi/.../mode_modules/.bin/lt.

2. Forward a local port specifying a subdomain:

lt --subdomain <subdomain_name> --port <port_number>
# example
lt --subdomain my-unique-subdomain-name --port 8080

How to publicly access a web app, running locally on RPI

1. Need a static IP address from the ISP with a specific port (commonly 80) open allowing incoming request.
2. Run the web app with node/rginx/apache2 locally in RPI.
3. If the RPI is connected with router Wifi, then forward the interal/localhost port to public. On TP-link, go to Advanced > Port Forwarding to map a port to that local port.

TroubleShooting

• Trouble: Problem connecting to RPI with VSCode remote explorer with previous ssh config.
  Shoot: Remove the previous ssh SHA1 string from the computer. On Windows, it is in os drive/users/<username>/.ssh/known_hosts.

• Trouble Where to look for system logs, crash/error/warning logs:
  Shoot:

  • The command dmesg will return most of the activity of the current boot. dmesg returns every event after the boot and how long after the boot, in seconds.
  • The files /var/log/messages /var/log/syslog and /var/log/kern.log will return pretty much every event you could ever need to know to figure out what happened.

Useful Notes

• Use node command to add a node.js app on device startup. Running with npm command will run an extra npm process.

• Installing Node.js with NodeSource will resolve the ARM architecture and other compatibilities and downloads the correct version of the Node.js distribution. So Node.js installed with one model Raspberry PI may not run on other model(s). Like RPI3 is ARMv71, so Node.js ARMv71 distribution has to be installed to run on that machine. But after that this installation will not run on RPI Zero W, which is ARMv61. ARMv61 distribution of Node.js has to be installed too.

• To change or add WIFI information diretly from the memory card: Wifi network information saved in /etc/wpa_supplicant/wpa_supplicant.conf. To add new network information, add following block of code at the end of the file:

  network={
     ssid="<ssid_name>"
     psk="<plain_text_password>"
  }
  

  IMPORTANT Don't use apostroph (') in wifi passwork.

Useful Linux terminal commands

• Update & Upgrade PI OS

  sudo apt-get update
  sudo apt-get upgrade
  

  If problem occurred on upgrade, run this:

  sudo apt-get upgrade --fix-missing
  

  • Firmware update command sudo rpi-eeprom-update only available in RPi-4.

• List current directory

  ls
  

• Make directory

  mkdir <dir_name>
  

• Removing files and directories

  • Remove non-empty directory:

  rm -r <dir_name>
  

  • Remove all files of current directory, with recursive (-r) and force (-f) arguments:

  rm -r -f *.*
  

  • To delete or skip one by one, type the following:

  rm -i mydir/*

  After each file name displays, type y and press Enter to delete the file. Or to keep the file, just press Enter.

• Size of current directory
  -s to display only the total size, -h to display sizes in a human-readable format.

  sudo du -sh
  

  Specific directory

  sudo du -sh /home/pi
  

• Open a file in terminal with root permission:

  sudo nano <file_name>
  

• Shutdown, reboot etc

  sudo shutdown -h now
  

  sudo reboot
  

• List running processes All processes

  ps -e
  

  Filtering processes

  ps -e | grep <partial_process_name>
  

  Example:

  ps -e | grep node

  Real-time process listing by CPU usage

  top
  

• How to run a process in the background, permanently

  nohup <command_and_arguments> &
  

  nohup is short for "no hang-up". Ending '&' will run the command in the background.

• Kill a process:

  kill -9 <process_id>
  # or
  kill -9 <process_name>

  • -9 denotes SIGKILL; to see full list of kill parameters, use kill -l

• Top 10 processes sorted by CPU usage. This will count the header line as one record, that's why -11

  ps -eo pmem,time,stat,pid,pcpu,command --sort -pcpu | head -11

• Memory status

  free -h

• Run Tailscale vpn

  sudo tailscale up
  

  • Seems like Tailscale doesn't support ARMv61 process (i.e. RPi Zero W)

• Raspberry PI OS configurations

  sudo rasp-config
  

• GPU temperature

  vcgencmd measure_temp

  Regular expression to get only the value:

  vcgencmd measure_temp | grep  -o -E '[[:digit:]].*'

• CPU Temperature

  cat /sys/class/thermal/thermal_zone0/temp

• Cpu Information

  cat /proc/cpuinfo

  Use grep to filter information:

  cat /proc/cpuinfo | grep Model

• Check under-voltage warning:

  dmesg | grep voltage

  Also Cpu throttle status, good value is 0x0:

  vcgencmd get_throttled

  Bit Position	Meaning
  0	Under-voltage detected
  1	Arm frequency capped
  2	Currently throttled
  3	Soft temperature limit active
  16	Under-voltage has occurred
  17	Arm frequency capping has occurred
  18	Throttling has occurred
  19	Soft temperature limit has occurred

  For example, a throlled value shows 0x80008. Now convert this hex value to binary in two steps - hex to decimal, decimal to binary.

  let n = parseInt('0x80000', 16) // must use quotes. without quotes it shows different value.
  n.toString(2) // result: 10000000000000000000

  Here, in the value '10000000000000000000' 19th bit is turned on which means soft temperature limit has occurred.

• Save any command output to a file <command_and_arguments> > <file_name>. Example:

  top -b -n1 > output.txt

  File will be saved in current directory.

• System log path /var/log/messages. Additionally enable watch dog to do vigorous logging.

• Download/copy over SSH:

  scp username@remotehost:file_name.ext /my/local/destination/path