by schnerd

Using headless Chrome as an automated screenshot tool (alternative to PhantomJS)

449 Stars 54 Forks Last release: Not found 41 Commits 1 Releases

Available items

No Items, yet!

The developer of this repository has not created any items for sale yet. Need a bug fixed? Help with integration? A different license? Create a request here:

May 2018 The Chrome team has released Puppeteer, which makes much of the code in this repository obsolete. As such, this project is no longer maintained and issues have been disabled. Please check out Puppeteer.

Using headless Chrome as an automated screenshot tool

This repo contains an example implementation of using headless Chrome as an automated screenshot tool on linux, which is a common use case for PhantomJS. Contributions are welcome.

Chrome Version

Headless Chrome is still new and unstable, and the API changes with each new major Chrome version. Our master branch is currently targeting Chrome 60. You may need to modify the script if you wish to target another version.

Setup on Linux

The setup below was used on a Vagrant running Ubuntu 14 Trusty Tahr. It assumes you've already cloned the repo and run

npm install
# Install Google Chrome
sudo apt-get install libxss1 libappindicator1 libindicator7
sudo dpkg -i google-chrome*.deb  # Might show "errors", fixed by next line
sudo apt-get install -f

Install Node Stable (v8)

curl -sL | sudo -E bash - sudo apt-get install -y nodejs

Run Chrome as background process

--disable-gpu currently required, see link above

google-chrome --headless --hide-scrollbars --remote-debugging-port=9222 --disable-gpu &

Take the screenshot

node index.js --url=""

The screenshot will then be available as output.png

Setup on OSX

Headless Chrome is still highly unstable on OSX (see issue #1). At this point in time I recommend just running chrome & node in docker or vagrant (Vagrantfile pull requests welcome).

If you must run it natively, use the following commands:

/Applications/Google\\ Chrome --headless --remote-debugging-port=9222 --disable-gpu
node index.js --url=""

If screenshots on Mac do not appear to be working, please report an issue on ChromeDevTools/devtools-protocol, cyrus-and/chrome-remote-interface, or chromium itself.

Setup on Docker

Build image:

docker build -t chrome-headless-screenshots-app .

Run container, saving output in local dir in the file output.png:

docker run -it -v ${PWD}:/var/output/ --rm chrome-headless-screenshots-app --url=""

Note that for newer versions of docker, you may see an error like:

libudev: udev_has_devtmpfs: name_to_handle_at on /dev: Operation not permitted

This appears to be harmless, and is caused by Chrome attempting to use /dev/shm for caching, which is more restricted in later versions of docker. If you really want to get rid of the warning then you can either dig a bit further to put together a seccomp file for chrome (see for a start), or use

--security-opt seccomp:unconfined
but note that this is not recommended ("seccomp is instrumental for running Docker containers with least privilege. It is not recommended to change the default seccomp profile."). For now, we recommend just living with the warning.

Other Resources

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.