I think adding the handling is the first thing to do, if this freaks out someone who knows what they are doing, what will it do to someone new to the project?
Also, maybe the drivers will need a quirks mode? On April 12, 2019 at 09:26:25, Christofer Dutz ([email protected]) wrote: Hi Otto, when inspecting the TCP traffic, for example Siemens LOGO devices tend to just hang up without sending a TCP ack if they get something they don't like. It's not TCP compliant, but it's the way they roll :-( So I know when trying to access a LOGO device with the S7 protocol, I had to disable the part where the driver tries to find out what type of PLC he is talking to because of this ... don't really know how we can counter-react this situation ... How about me adding the Exception handler and simply logging things there so we can define a breakpoint and try to find out what's going on? Chris Am 12.04.19, 15:06 schrieb "Otto Fowler" <[email protected]>: I would think you’d want to handle it, and then investigate the reason for it in this instance On April 12, 2019 at 08:51:05, Christofer Dutz ([email protected]) wrote: Hi Otto, I think the English translation would be something like: "Connection reset by remote" or "Connection reset by peer" Not sure if this helps understand what's going on ... I always thought it was related to the PLC hanging up without following the correct procedure for doing that. Sort of like just hanging up the phone without saying goodbye. Perhaps really implementing an exception handler and just swallowing the exception Because we know we're operating in "unpolite" environments ... but not sure what We could do to gracefully handle the situation. @Julian ... would it be ok to catch that and simply log an "INFO" severity message ("Remote hung up unexpectedly") Chris Am 12.04.19, 13:52 schrieb "Otto Fowler" <[email protected]>: The netty handlers don’t implement exceptionCaught so the default is to spew to logs. I don’t speak german so I don’t know what the exception message is. Maybe the handlers should implement the method. On April 12, 2019 at 04:56:40, Julian Feinauer ([email protected]) wrote: An exceptionCaught() event was fired, and it reached at the tail of the pipeline. It usually means the last handler in the pipeline did not handle the exception. java.io.IOException
