Lots of WARNING logs

[bachya]

Submitting on behalf of an issue discovered by the use of this library (and python-socketio) in Home Assistant: home-assistant/core#22295

We are receiving lots of these WARNING-level entries in our logs:

2019-03-22 12:07:38 WARNING (MainThread) [engineio.client] Read loop: WebSocket connection was closed, aborting
2019-03-22 12:07:38 WARNING (MainThread) [engineio.client] Write loop: WebSocket connection was closed, aborting

From my understanding (and previous work with you), I expect these occur when the client gets unexpectedly disconnected from the socket. Since the client automatically reconnects, there isn't any user-facing impact – they merely see large amounts of "nothing to worry about" logs.

Would you consider taking this to INFO level if a reconnect is successful? Open to other alternatives and thoughts.

[miguelgrinberg]

I'm okay with downgrading these messages to INFO, that seems reasonable. Do you wan to create a PR for this change?

[bachya]

Will do! Is there other logging – here or in python-socketio – that you feel falls under the same banner (i.e., because the end-result was taken care of, WARNING is too steep)? If so, happy to adjust via PR, as well.