Hello,
I am setting up a distributed test system, with the docker images selenium/hub and selenium/node-chrome. There will be multiple node-chrome images, all launched on different hardware. I want to launch a specific tests on specific hardware (specific node-chrome instance).
Online I found multiple discussions / explanations about setting browser capabilities, like
here . From that report, setting he environment variable NODE_APPLICATION_NAME seems the most easy way. Especially because it seems the only way to avoid re-specifying all the capabilities.
But, I also found in an issue
here that this application name is deliberately not supported anymore from grid4 onwards.
So, what is currently the recommended way to set a specific capability on a chrome node? I prefer to not create my own Dockerfile, but just use a docker-compose that launches the standard selenium/node-chrome:latest image with any options/env vars/config files/... that are needed (without having to specify/overwrite/... standard default config options)
Kind regards,
Pieter-Jan