chrome-headless | A simple , practical Chrome headless script example | Automation library

Use https://webrtc.github.io/samples/src/content/peerconnection/trickle-ice/ to gather a srvflx candicate.

Cant receive srvflx candicate on Ubuntu Server(Linux). Works on Windows.

• sudo apt-get update && apt-get upgrade
• sudo ufw disable
• curl ipecho.net/plain // Returns a pingable ip

• curl -sL https://deb.nodesource.com/setup_13.x | sudo -E bash - && sudo apt-get install -y nodejs
• mkdir -p /tmp/puppeteertest && cd /tmp/puppeteertest && npm init
• npm i puppeteer --save
• Install all libs listed here https://github.com/puppeteer/puppeteer/blob/master/docs/troubleshooting.md#chrome-headless-doesnt-launch-on-unix
  • sudo apt-get install -y gconf-service libasound2 libatk1.0-0 libatk-bridge2.0-0 libc6 libcairo2 libcups2 libdbus-1-3 libexpat1 libfontconfig1 libgcc1 libgconf-2-4 libgdk-pixbuf2.0-0 libglib2.0-0 libgtk-3-0 libnspr4 libpango-1.0-0 libpangocairo-1.0-0 libstdc++6 libx11-6 libx11-xcb1 libxcb1 libxcomposite1 libxcursor1 libxdamage1 libxext6 libxfixes3 libxi6 libxrandr2 libxrender1 libxss1 libxtst6 ca-certificates fonts-liberation libappindicator1 libnss3 lsb-release xdg-utils wget
• vi index.js // Create index.js with code below

I'm trying to select and console.log() the NodeList of all the links of a website in the terminal. However I'm unable to do so accessing certain websites - google.com, facebook.com, instagram.com.

I know that elements are there, because I can certainly log them in the actual Chromium console, which loads separately, using document.querySelectorAll('a'). But when I'm trying to extract and log links in Node terminal, using

UPDATE

As @Joaquín said, the base image was lacking dependencies. After a lot of research along with trial and error, I settled on using the heavy servercore image along with the necessary fonts required by Chrome. You can find my test app and Dockerfile on my github.

I was able to get it running in Azure App Service as a Windows container, but the image is many times larger and much slower startup time than my Linux version. I'm still very new to Docker containers, so it's been interesting so far.

Running my puppeteer app in a Windows container in Azure App Services ends up with UnhandledPromiseRejectionWarning: Error: Failed to launch chrome!

Yet, it works flawlessly in a Linux container in Azure App Services. Do I have to do something particularly special to get it to work in a Windows container? Or is it futile due to the sandboxing limitations in a Windows app service (container or not)? Though, if this were the case, then I would've expected a spawn UNKNOWN error...

Some of the things I've tried included using the following puppeteer.launch() options:

1. ignoreDefaultArgs: ['--disable-extensions'] (per troubleshooting)

2. executablePath: ''

Here's the error and stack trace

We have used phantomjs on previous days for generating pdf with server-side prince tool. Nowadays we like to change the headless browser from phantomjs to another one.

I research about which one is quickly rendering both phantomjs and puppeteer. In my case Reporting app to generate the PDF, only I test it with puppeteer and phantomjs. phantomjs quickly process the HTML compare to puppeteer.

And then I learn google-chrome-headless options to generate the PDF. It looks like a Chrome browser GUI application built-in modules. I am working with nodejs application. I am used with the below command from the command line

I'm trying to make my online bot undetectable. I read number of articles how to do it and I took all tips together and used them. One of them is to change window.navigator.webdriver.

I managed to change window.navigator.webdriver within puppeteer by this code:

I'm trying to export an image rendered with WebGL on a linux server without a GPU. To do this I'm using headless Chrome however the exported image is black (example exported image, taking a screenshot of page shows its just canvas that is black). I was hoping for some help figuring out why this is happening.

To export the image I render the image into a canvas, export data via canvas.toDataURL('image/jpeg') and then post the data to the server. I'm using Pixi.js for rendering, if I use canvas renderer then everything works on the server; It's WebGL rendering thats not working. It's worth noting the WebGL render works fine in Chrome 63 on a Macbook.

To control Chrome I'm using Puppeteer. All I'm doing is opening a page, waiting a second, and then closing it again:

I have this gitlab-ci.yml to build my SpringBoot app:

Trying to run chrome headless as a container (Image: https://hub.docker.com/r/alpeware/chrome-headless-trunk) in my docker-compose and connecting to it from another container.

To actually connect to chrome inside the container, I first need to retrieve the webSocketDebuggerUrl, which is available at http://0.0.0.0:9222/json/version of the chrome-headless.

The Problem is: my request to this path always fails with

RequestError: Error: connect ECONNREFUSED 0.0.0.0:9222

and cant get the webSocketDebuggerUrl to connect to chrome.

Also if I visit http://0.0.0.0:9222/json/version in my browser myself, copy the url and hardcode it into my puppeteer.connect(), it ONLY works as expected, if I replace the address of '0.0.0.0' to my (linked) container-name (specified in docker-compose): http://chrome:9222/json/version

If I try to request the webSocketDebuggerUrl from /json/version while using container-name address (http://chrome:9222/json/version) i get the error

StatusCodeError: 500 - "Host header is specified and is not an IP address or localhost."

I'm trying to zip my dist folder and post to it Azure's kudu zipdeploy endpoint but I keep getting this error regardless of the fact that I have installed a zipping package myself.

My docker image: weltn24/up-docker-node-chrome-headless on tag: chrome-66_node-10.0_yarn-1.6.0

The part that fails in my yml file: