Hi, We just tried running yahns on production, and would like to give some feedback. Eventually we turned back to Unicorn, but that's because apparently Rails, or to be specific, ActiveRecord (4.1.10) is not thread safe and giving random errors once in a while.
However, we're also trying to split out an API server which is not running on Rails, therefore we should be able to use a threaded server. During the experiments, we found that whenever a worker thread died due to LoadError raised from the application, which is not a StandardError therefore was not rescued at all, crashing the worker thread (assumed, not verified). When this happened, the client just hanged forever with yahns. Is there something we can do about this? Would yahns respawn a new worker thread? Can we close the socket when this happen? I am aware that yahns is *extremely sensitive to fatal bugs in the applications it hosts*, so I am just curious. For reference, Puma would immediately close the socket without sending anything, and Unicorn would log the error backtrace and kill the worker (If I read it correctly). In this case, Unicorn helped me figure out what's happened. Cheers,
