headless-chrome | Headless chrome using Puppeteer that can run on Heroku | UI Testing library

The solution is actually simple.

I found a workaround by using export headless=0; behave in the non-headless bash script and it worked.

The caveat to this method is that I have to use os.getenv('headless') which refers to the value set in my .env file. that was headless=1.

The final webconfig.py looks like this.

You dont need a seperate Node / NPM Container. The Browsershot is part of your applications core. To make it work you need nodejs and npm. That saying I would construct my own container image incuding PHP and Node. That makes the implementation of Browershot more simple.

If neede you can split them up later. But in that case I would create a new BrowserShot service (PHP + Node).

As you can see, there are multiple ways doing it.

What about using NGINX Unit for your applications container base? You could use the Unit PHP base image and add node on top of it.

I did that here:

https://github.com/nginx/unit-examples/blob/master/Dockerfile

I found a solution for the above problem statement.

First I tried using chrome and firefox instead of chromium. But apline doesn't had chrome and so switched my base image to ubuntu. Also, in general, ubuntu is suggested [Reference: https://pythonspeed.com/articles/base-image-python-docker-images/] as a best docker base image for running Python Applications.

But even after changing to ubuntu as new docker base image with chrome and firefox, it is the same error (blank page white screen).

Below error as well,

Gunicorn is an application server for running your python application instance, it is a pure-Python HTTP server for WSGI applications. It allows you to run any Python application concurrently by running multiple Python processes within a single dyno.

Please have a look into the following Tutorial which explains in detail regarding gunicorn.

Regarding Cloud Run, to deploy to Cloud Run, please follow next steps or the Cloud Run Official Documentation:

1) Create a folder

2) In that folder, create a file named main.py and write your Flask code

Example of simple Flask code

I hit exactly the same wall as you and the same garbled response. I thought it was a bug - even after reboots it didn't come back so i was quietly waiting for a patch.

I just had a more detailed look now and - AT A GUESS - it seems to be related to the open chrome instances.... When i launch the new chrome with the remote debugging tag it attached to an existing chrome not in its own browser.

What i did:

Run these two commands to close all your chrome/chromedriver instances (including this one if you're using chrome - so use something else or copy it to notepad):

yeah i think you could add a post-crawling step in the evaluatePage callback applying your code:

This is it.

After 2 or 3 days struggling with this and trying to find the exact problem, it finally works by disabling the anti-aliasing in unity, which seems to throw some OpenGL errors.

That said, the Unity3D WebGL rendering works headlessly (through chromium and puppeteer) even without swift shader, though that is probably what you want for no-gpu scenarios (e.g., some server).

The log output, though still weird, for both scenarios (with and without swift shader) is the following (note that the framebuffer incomplete errors are gone):

Anyway, I ended up using PageSpeed Insights API instead: https://developers.google.com/speed/docs/insights/v5/get-started

It produces the same JSON that Lighthouse CLI does, but you don't need to setup Chrome or anything, it's just a HTTP request.