Currently HttpShardHandlerFactory and UpdateShardHandler default to infinite timeouts for socket connection and read. This can lead to undesirable behaviour, for example, if a machine crashes, then searches in progress will wait forever for a result to come back and end up using threads which will only get terminated at shutdown.
We should have some finite default, however conservative it might be. These parameters are already configurable, so for expert uses, they can be increased if necessary anyway.
Will attach a patch to set connection timeout to 60s and read timeout to 600s, but I am not too concerned about the actual value as long as there is one.