I have 3 browsers and up to 160 asyncio tasks. These tasks are split equally among the browsers, and I have a total of 40 browser contexts (split equally among the browsers). Sometimes, when calling 'goto', I get "Connection closed while reading from the driver" error, however, this makes all other contexts to start failing. When trying to clean up, I still get some driver errors: "pipe closed by peer or os.write(pipe, data) raised exception.".
Upon restarting the workflow, if I revisit the same URLs again, the application gets into a crash loop with the same errors. If I change the URLs, it works fine (however, revisiting the URLs later may also work fine, so it is not a particular URL that is causing this issue). I have no clue what is causing this, so I would appreciate any suggestions or at least ideas. Is this a bug from my app or Playwright?
This thread is trying to answer question "What is causing the 'Connection closed while reading from the driver' error and subsequent crash loop in Playwright Python when running multiple asyncio tasks across different browser contexts?"
Rayrun is a community for QA engineers. I am constantly looking for new ways to add value to people learning Playwright and other browser automation frameworks. If you have feedback, email [email protected].