Infinite loop in _connect_attempt() - (class ConnectionHandler in protocol/connection.py)

[robertmyhill]

In _connect_attempt(), the "while not close_connection:" loop sometimes loops forever, pausing about 7 seconds between each loop.

I can reproduce this, usually within 5 minutes, in a three-node zookeeper cluster by continually restarting each of the zookeeper services.

I think the issue is that the connection to the server has gone away, and the connection loss is not being detected in the "while not close_connection:" loop.

In the "while not close_connection" loop, neither of the calls to _send_ping() nor _send_request() are raising exceptions. Would you expect them to do so? Could it be that I don't have TCP configured correctly?

Thanks for any help/thought on this.

Robert