I know this is sort of counter to the purpose of headless automation, but...
I've got an automation test running using Selenium and Chromedriver in headless mode. I'd prefer to keep it running headless, but occasionally, it runs into an error that really needs to be looked at and interacted with. Is it possible to render and interact with a headless session? Maybe by duplicating the headless browser in a non-headless one? I can connect through remote-debugging, but the Dev Tools doesn't seem to do allow me to view the rendered page or interact with anything.
I am able to take screenshots, which sort of helps. But I'm really looking for the ability to interact--there's some drag-and-drop elements that aren't working well with Selenium that are causing issues occasionally.