Raspberry Pi Kiosk Using Chromium - Pi My Life Up
Raspberry Pi Kiosk Using Chromium - Pi My Life Up
Raspberry Pi Kiosk Using Chromium - Pi My Life Up
-37%
R$1.799
R$863,89 R$1.124,10
-28%
R$1.454,65 R$659
R$1.396,61 R$471,11
We utilize Chromium as it is one of the best-supported web browsers and openly supports the functionality to act in a
kiosk mode.
It is also easy to control through key presses which we can simulate using the xdotool that we install during this guide.
This guide will give you a good idea on how to use autologin tasks and simple bash scripts to perform numerous tasks.
l
close
Please note to do this tutorial you will need to be running the full version of Raspbian as this relies on the GUI that
comes with it to properly operate.
Equipment List
The entire list of all the pieces of equipment that we made use of for this Raspberry Pi Kiosk tutorial is listed below.
Recommended
Raspberry Pi 2 or 3
Micro SD Card
Power Supply
TV or Monitor
HDMI Cord
Optional
Raspberry Pi Case
Video
In this video, we show you the process of setting up your Raspberry Pi to operate as a Chromium powered kiosk as well
as showing you how to get it to start at boot.
If you prefer a written and more thorough explanation, then check out our written guide below.
l
close
1. Before we get started with this tutorial, we will be first removing some packages that we don’t need for our Raspberry
Pi kiosk.
Removing these packages will free up some much-needed memory and reduce the number of packages that will be
updated every time you update your Raspberry Pi.
To do this just run the following three commands on your Raspberry Pi. We have split these into three different
commands to make them easier to copy and write out.
2. Now that we have removed these numerous packages from our Raspberry Pi Kiosk we will need to run some cleanup
commands.
To remove any unnecessary lingering packages and to clean out the local repository of retrieved packages run the
following two commands on your Raspberry Pi.
3. Now that we have removed the bulk of the unnecessary applications, we must now make sure that our installation of
Raspbian is up to date. Also, make sure you have SSH enabled as it is very handy if you need to edit any files later on.
We can use the following two commands within the terminal on our Raspberry Pi to update all the packages to their
latest versions.
4. We now need also to install xdotool. This tool will allow our bash script to execute key presses without anyone being
on the device. We also install the unclutter package, this will enable us to hide the mouse from the display.
Just run the following command on your Raspberry Pi to install the package.
l
close
5. Now that those packages have been installed we can now proceed to the next stage of this tutorial. That is setting up
Raspbian to auto login to our user. Having to log in every time for a kiosk would be an annoyance.
Desktop autologin is the default behavior but if for some reason you have changed it follow the next few steps to switch
it back. Otherwise, skip to step 6 of this tutorial.
5a. Run the following command on your Raspberry Pi to load up the Raspi-config tool. We will be using this tool to
enable auto login.
sudo raspi-config
5b. Now within the tool go to 3 Boot Options -> B1 Desktop / CLI -> B4 Desktop Autologin
5c. Desktop autologin should now be enabled and you can safely quit out of the raspi-config tool.
6. Now that we have enabled desktop autologin we need to go ahead and write our kiosk.sh script.
Begin writing our kiosk bash script by running the following command on the Raspberry Pi.
nano /home/pi/kiosk.sh
1. Within this file enter the following lines of code, we will explain the important parts of the script so you can better
bend it to your needs.
#!/bin/bash
The very first line defines what the command line interpreter (CLI) should use to try and execute this file. This is useful for
cases where you don’t want to have to specify the specific application required every time you run the script.
xset s noblank
xset s off
xset -dpms
These three lines are pretty important as they help us stop the Raspberry Pi’s display power management system from
kicking in and blanking out the screen.
l
Basically, these three commands set the current xsession not to blank out the screensaver and then disables the close
screensaver altogether. The third line disables the entire “display power management system” meaning that the desktop
interface should never blank out the screen.
This line runs the program we installed earlier called unclutter. This application will hide the mouse from the display
whenever it has been idle for longer then 0.5 seconds and will remove it even if it is over the root background.
You can adjust the idle timer to the number of seconds you want with each decimal place being a fraction of a second.
If you would prefer to remove the mouse instantly, then remove the -idle 0.5 from the command.
sed -i 's/"exited_cleanly":false/"exited_cleanly":true/'
/home/pi/.config/chromium/Default/Preferences
sed -i 's/"exit_type":"Crashed"/"exit_type":"Normal"/'
/home/pi/.config/chromium/Default/Preferences
These two lines of the script use sed to search through the Chromium preferences file and clear out any flags that would
make the warning bar appear, a behavior you don’t really want happening on your Raspberry Pi Kiosk.
If Chromium ever crashes or is closed suddenly, the lines above will ensure you don’t have to get hold of a mouse and
keyboard to clear the warning bar that would typically appear at the top of the browser.
This line launches Chromium with our parameters. We will go through each of these parameters so you know what you
can modify, and how you can modify it.
--kiosk
This flag sets Chromium to operate in Kiosk mode, this locks it into a particular set of features and only allows limited
access to both the web browser and any other OS functionality.
Chromium’s kiosk functionality takes full control of the screen, maximizing Chromium to the full size of your screen and
stops user input from being accepted by the OS, effectively trapping the end user within a sandbox.
--noerrdialogs
This option tells Chromium that it should not display any of its error dialogs to the end user.
It is crucial if you don’t want the end user to know if anything has or is going wrong with Chromium, this goes alongside
our code to clear the “exited_cleanly” and “exit_type” state earlier in the code.
--disable-infobars
We use this to disable Chromium from displaying its info bar to the end user.
The info bar can be used by Chromium to notify them of certain things such as that Chromium is not their default web
browser. Of course, as we are using this as a kiosk, we don’t need the user to know any information that Chromium ×
might want to display.
l
close
https://pimylifeup.com https://www.adafruit.com
These are the two web pages that the script will open, each will be opened in a new tab. You can add additional web
pages/tabs by adding to this list by separating each one with a space.
while true; do
xdotool keydown ctrl+Tab; xdotool keyup ctrl+Tab;
sleep 15
done
These lines run a very simple infinite while loop that uses xdotool to mimic pressing Ctrl + Tab. Making Chromium
switch to the next tab.
After xdotool has executed its key presses, it then puts the loop to sleep for 15 seconds. To change how long the loop
sleeps for before it executes the xdotool command again just change the sleep 15 command.
You can also use this method to add a screen refresh, this may be important when you want to display live scores. The
command for this should look something like what we have shown below.
2. Once you have entered all the code for our Raspberry Pi kiosk script, it should look somewhat similar to what we have
displayed below.
#!/bin/bash
xset s noblank
xset s off
xset -dpms
sed -i 's/"exited_cleanly":false/"exited_cleanly":true/'
/home/pi/.config/chromium/Default/Preferences
sed -i 's/"exit_type":"Crashed"/"exit_type":"Normal"/'
/home/pi/.config/chromium/Default/Preferences
while true; do
xdotool keydown ctrl+Tab; xdotool keyup ctrl+Tab;
sleep 10
done
3. Once you are sure everything is correct, save the file by pressing CTRL + X then Y and finally ENTER.
l
close
1. Before we get started we need to first utilize the following command to work out what the current display value is.
This value is used for the operating system to know what screen to display the Chromium kiosk to, without it the kiosk
will either fail to load or load up on the incorrect screen.
Run the following command to echo out the value of the “$DISPLAY” system variable.
echo $DISPLAY
Make sure you remember this value as you may need it in step 3 of this section.
2. To get our Raspberry Pi Kiosk to start at boot we will need to go ahead and create a service file by running the
command below.
This service file will tell the operating system what file we want to be executed as well as that we want the GUI to be
available before starting up the software.
3. Within our kiosk service file, enter the following lines of text.
These lines are what will define our service kiosk service, and that we want to run our kiosk.sh script when the system
loads into the operating system.
While entering these lines you may have to modify the “Environment=DISPLAY=:” line, replacing the “0” with the value
that you retrieved from the command you used in step 1 of this section.
[Unit]
Description=Chromium Kiosk
Wants=graphical.target
After=graphical.target
[Service]
Environment=DISPLAY=:0.0
Environment=XAUTHORITY=/home/pi/.Xauthority
Type=simple
ExecStart=/bin/bash /home/pi/kiosk.sh
Restart=on-abort
User=pi
Group=pi
[Install]
WantedBy=graphical.target
Once you have entered everything into the file, save the file by pressing CTRL + X followed by Y then ENTER.
4. Now that we have created the service file for our Raspberry Pi Kiosk we can go ahead and now enable it by running
the following command.
By enabling the service, we will allow our Chromium Kiosk to start up at boot automatically and will enable the systemd ×
to service manager to monitor it.
l
close
sudo systemctl enable kiosk.service
5. With the Kiosk service now enabled you can either choose to restart the Raspberry Pi or start the service now by
running the following command.
6. If you ever want to check the status of your Raspberry Pi Kiosk’s service, you can run the command below.
This command will return various information about the service, including previously returned lines from the software
which can help you debug what’s going wrong when the service is failed.
If this command shows the status as “Active: active (running)” then everything is now working as it should be, and your
Raspberry Pi Chromium Kiosk should be up and operating correctly.
7. Now with everything up and running correctly, if there is for any reason you want to stop the service from running,
you can utilize the following command.
By stopping the kiosk service, the service manager will kill off all processes associated with it. This command will stop our
kiosk.sh script from running while also terminating the open Chromium browser.
8. Finally, if you ever want to disable your Kiosk, you can utilize the following command.
This command will stop the Kiosk service from running on boot until you re-enable it.
1. One thing you might want to do is to enforce the resolution that the Raspberry Pi is going to use. Setting the
resolution can be quite handy as the Raspberry Pi’s built-in detection can sometimes be a bit flakey.
To begin setting the resolution, we must first load up the Raspberry Pi configuration tool by running the following
command.
sudo raspi-config
2. Within the configuration tool, you will want to start by going to “7 Advanced Options“.
×
l
close
3. Now that we are in the advanced options section you should see an option labeled “A5 Resolution” select that
option.
4. Within here find and select the resolution that best fits your screen and press ENTER.
5. With your resolution now set you will need to restart the Raspberry Pi. Do this by first exiting out of the configuration
tool by pressing ESC, then entering the following command into the Raspberry Pi’s terminal.
sudo reboot
6. Your Raspberry Pi should now restart and be running at the specified resolution.
There are so many ways you can extend this tutorial. For example, you can set up a web server on the Raspberry Pi and
have it serve local web pages to be displayed on your Kiosk.
It’s perfect if you want the results of a competition or basically any other sort of information you want to be displayed.
By now you should have your Raspberry Pi successfully booting into the Kiosk mode of Chromium. If you have any issues
with this Raspberry Pi kiosk tutorial or want to leave feedback, then feel free to leave a comment below.
RasPad Review: The Freedom to Arduino Accelerometer using the Basic Guide to Voltage Dividers
Develop ADXL345
How to Install VLC for the Raspberry Using IFTTT with the Raspberry Pi Raspberry Pi DNS Settings: How to
Pi Change the DNS
email Sign up »
l
close
77 Comments
Reply
Hi,
I just ran through the tutorial to verify everythings still working and the service succesfully enabled on my Raspberry
Pi.
[Install]
WantedBy=graphical.target
Cheers,
Emmet
Reply
You can get away with 1gb, but more is always better.
Thank you so much! I needed a way to display a Node-red dashboard on boot, I tried several other methods (wasting a
better part of the day) and this is the first one to work! ×
l
close
Reply
After spending more time than I care to admit, this worked the first time, right out of the box! I needed to set up a
PI to run Plex for the family room, so that it is as bullet-proof as possible! The Plex server is on one PC server in the
computer room, and the File server is on another, also in the server room, so that files can’t be added/deleted/ or
changed without the admin (me) doing it from a controlled environment. Thanks for such a WONDERFUL tutorial!
Also, another tip: For those using a RPi 4, the HDMI audio ONLY plays thru HDMI 0, the one on the left. That one bit
me for a LONG time!
Hi, I cant wait to try this. I am curious if you could let me know how to refresh the browser once every hr or whatever I set
it to. ie 1 minute or 24 hrs.
I would like to use this to display a google slides presentation as a messaging display at work and would like it to refresh
the slide/message without having to manually reload the page.
Reply
Let’s say you have 2 different tabs, and you want to play for 60 seconds, but the other 120 seconds before switching.
How would that be scripted?
Reply
Hi Matt,
l
The code below should do what you’re after. It will display the first tab for 60 seconds then switch to the next tabclose
for 120 seconds.
You actually want something a little different in-case the tabs get messed up
Below will switch to the first tab in Chrome, wait 60 seconds, then go to the 2nd tab for 120 seconds you can do this
with multiple tabs in any order
Hi Emmet,
I must say your tutorial is awesome. But I would like to know if Chromium running in kiosk mode and there is not
keyboard and mouse attached to pi screen attached to pi can support touch but when chromium is running if any form
fill needs to happen on web how can we invoke matchbox-keyboard.
I will wait to hear from you.
Much appreciated.
Reply
Hi Mansoor,
I don’t have a touch screen to test this with but could you try adding “–touch-events” to the arguments list for
chromium.
Cheers,
Emmet
l
Emmet, this has been a game-changer for me, thank you so much for taking the time to publish this. I’m, however, close
having some issues with scrolling through the tabs in Chrome, I’ve followed the instructions and have this is in kiosk.sh
file:
while true; do
xdotool keydown ctrl+Tab; xdotool keyup ctrl+tab
sleep 10
xdotool keydown ctrl+r; xdotool keyup ctrl+r
sleep 20
done
but it doesn’t seem to be working? If you can help, I’d be hugely appreciative
Reply
Just to clarify, this is only a problem when in full screen mode. It works perfectly well outside of full screen…
Hi Liam,
This definitely does seem peculiar, while I couldn’t reproduce this behaviour myself I have tried to come up with a
solution.
Can you please try replacing the “while” loop with the block below. The addition i made here was to make xdotool
search for the chromium browser and force it in to focus.
while true; do
xdotool search --onlyvisible --class Chromium windowfocus
xdotool keydown ctrl+Tab; xdotool keyup ctrl+Tab;
sleep 10
done
Cheers,
Emmet
Hello, and thanks for the script. My problem is that it will not open/start since (failed = exit-code, code-exited status
1/failure)
The program is loaded but not active. Have no ideas, unless time settings are involved. My settings are CEST+1 (Sweden)
×
Reply
l
close
Kjell Gustafsson on June 10, 2019 at 2:40 pm
Reply to myself; check your display with echo $DISPLAY and use the value it generates. In my case it was 0.0
Happy ending 😉
Reply
Hi, what a great tutorial. Thank you so much! It’s working great and I just wanted to add that it works on folders as well.
I’m hardly a programmer or anything but replacing the the URLs with
/home/username/foldername/*.jpg
One question. This might be outside the scope of this .sh but is there a way to set a auto on/off? For instance: having it
start the kiosk at 9AM and powering off the kiosk at midnight (local time of course) without shutting the system down?
Just turning off the service and powering off the screen.
Or even just powering off at a specific time would be great. I tinkered with it but couldn’t figure out how.
Reply
Hi Shae.Da,
In answer to your question, one of the easiest ways to handle this would probably be to just use a crontab to stop
and start the device.
Adding the following lines to the crontab should work but i havent tested this myself.
sudo crontab -e
Reply
Hi Boris,
If you have removed the section that changes the tabs can you please try removing the & symbol thats at the end
of the chromium-browser call and let me know whether that resolves your issues.
Cheers,
Emmet
Script didn’t work for me out of the box because of this error: https://askubuntu.com/questions/476036/xset-unable-to-
open-display
You may want to update the docs to include that for some cases you need to add the following line after #!/bin/bash in
kiosk.sh:
export DISPLAY=:0
×
Reply
l
close
Emmet on April 27, 2019 at 6:57 pm
Hi Brian,
I see a few people are seeming to run into similar issues with the display not working as it should be which is
bizzare.
We will likely add a troubleshooting section to reflect the potential need to add export DISPLAY=:0 to the script but
just want to gather some more information on it.
Cheers,
Emmet
Reply
Hi Mike,
Apologies for taking a few days to reply, I went and looked into this and you may be able to avoid relying on an
extention by passing it few as an argument.
And change it so it looks something like what we have below. You can specify your own user agent of course, we are
just using an example one we grabbed off of Firefox’s developer webpage.
Let me know if this works out for you, if it works as it should we will go ahead and add a section to the tutorial on
this as I reckon others may need this to.
Cheers,
Emmet
Page 2 of 2
×
Previous 1 2
l
close
Leave a Reply
Your email address will not be published. Required fields are marked *
Comment
Name *
Email *
Post Comment
Trending
How to Setup a Raspberry Pi DNS Server Raspberry Pi Vivaldi: How to Setup the Vivaldi Browser
Raspberry Pi IRC Server: Setup your own Chat Network Arduino Traffic Light Project
l
close
l
close
l
close
l
close
l
close
l
close
l
close
l
close
l
close
l
close
l
close
l
close
l
close
l
close
l
close
l
close
l
close
Back to Top ↑
© 2020 Pi My Life Up