XHR Polling Breaks on gunicorn with Multiple Workers
When running gevent-socketio applications under gunicorn with multiple worker processes, the xhr-polling and jsonp-polling transports seem to fail after a short while.
I haven't dug into the code to be sure, but I'm assuming that there is some connection state being stored in memory, which is not available across processes. When gunicorn routes a client's request to a different worker process, the other process doesn't have access to the state, and the request fails.
I don't know if there's any good way to fix this, but it would be nice if it were at least stated explicitly in the documentation for using gunicorn with gevent-socketio, so future users won't spend so much time banging their head as we did.
https://github.com/abourget/gevent-socketio/issues/112#issuecomment-20478209
Please read invitation to Wednesday September 18th's sprint: https://groups.google.com/forum/#!topic/gevent-socketio/2OIRKA8M2uE