I’m wondering whether this is really the job that produces that log?
In your deploy:dev job you are running on a ruby:latest Docker container (which overrides image:node:16.14.2). You shouldn’t need to install ruby-dev in your before_script for that job, because your image is a Ruby image. You should just be able to say:
before_script:
- gem install dpl
Then in your after_script, you don’t have a line that says npm i npx that you show in the log?
Should be for creating variables, I’m not sure that you can rename them like this. It would be much easier to remove the variables section and change your script: