So I've added some logging into the messageSent and it's quite interesting although not terrible helpful. For on the messageSent event isn't fired on the first packet sent till after I've received the response from the client to it.
15:00:21.130 [NioProcessor-1] INFO o.a.m.filter.logging.LoggingFilter - CREATED 15:00:21.213 [NioProcessor-1] INFO c.s.m.p.s.MyEncoder - Sending: [10, 53, 46, 53, 46, 50, 45, 109, 50, 0, 0, 0, 0, 0, 112, 119, 112, 119, 112, 119, 112, 119, 0, 1, 130, 33, 0, 0, 8, 0, 12, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 109, 121, 115, 113, 108, 95, 100, 117, 109, 109, 121, 95, 112, 108, 117, 103, 105, 110, 0] 15:00:21.216 [NioProcessor-1] INFO o.a.m.filter.logging.LoggingFilter - OPENED 15:00:21.220 [NioProcessor-1] INFO o.a.m.filter.logging.LoggingFilter - SENT: HeapBuffer[pos=0 lim=76 cap=128: 48 00 00 00 0A 35 2E 35 2E 32 2D 6D 32 00 00 00...] 15:00:21.223 [NioProcessor-1] INFO o.a.m.filter.logging.LoggingFilter - RECEIVED: HeapBuffer[pos=0 lim=66 cap=2048: 3E 00 00 01 85 A6 7F 00 00 00 00 01 04 00 00 00...] 15:00:21.224 [NioProcessor-1] DEBUG o.a.m.f.codec.ProtocolCodecFilter - Processing a MESSAGE_RECEIVED for session 1 15:00:21.236 [NioProcessor-1] DEBUG c.s.m.p.s.MyDecoder - Decoding packet 1 with length 62 15:00:21.236 [NioProcessor-1] DEBUG c.s.m.p.s.MyHandler - Packet recieved 15:00:21.236 [NioProcessor-1] DEBUG c.s.m.p.s.MyHandler - Handshake response 15:00:29.893 [NioProcessor-1] DEBUG c.s.m.p.s.MyHandler - Send request 15:00:29.895 [NioProcessor-1] INFO c.s.m.p.s.MyEncoder - Sending: [254, 109, 121, 115, 113, 108, 95, 110, 97, 116, 105, 118, 101, 95, 112, 97, 115, 115, 119, 111, 114, 100, 0, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 0] 15:00:29.895 [NioProcessor-1] INFO o.a.m.filter.logging.LoggingFilter - SENT: HeapBuffer[pos=0 lim=0 cap=0: empty] 15:00:29.896 [NioProcessor-1] DEBUG c.s.m.p.s.MyHandler - Message sent: [10, 53, 46, 53, 46, 50, 45, 109, 50, 0, 0, 0, 0, 0, 112, 119, 112, 119, 112, 119, 112, 119, 0, 1, 130, 33, 0, 0, 8, 0, 12, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 109, 121, 115, 113, 108, 95, 100, 117, 109, 109, 121, 95, 112, 108, 117, 103, 105, 110, 0] 15:00:29.897 [NioProcessor-1] INFO o.a.m.filter.logging.LoggingFilter - SENT: HeapBuffer[pos=0 lim=48 cap=64: 2C 00 00 00 FE 6D 79 73 71 6C 5F 6E 61 74 69 76...] 15:00:29.898 [NioProcessor-1] INFO o.a.m.filter.logging.LoggingFilter - SENT: HeapBuffer[pos=0 lim=0 cap=0: empty] 15:00:29.898 [NioProcessor-1] DEBUG c.s.m.p.s.MyHandler - Message sent: [254, 109, 121, 115, 113, 108, 95, 110, 97, 116, 105, 118, 101, 95, 112, 97, 115, 115, 119, 111, 114, 100, 0, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 112, 119, 0] 15:00:29.900 [NioProcessor-1] INFO o.a.m.filter.logging.LoggingFilter - CLOSED Sorry I can't show any more code than in the original message (although I think I have everything Mina related there) as it's confidential. Jennifer On 3 April 2013 20:58, Emmanuel Lécharny <[email protected]> wrote: > Le 4/3/13 9:50 PM, Jennifer Cumming a écrit : > > The client is reporting an error that implies the server either closed > the > > connection or timed out. As I can see the packet being written in the > > logging filter well before the timeout expires > Are you receiving a messageSent event ? If not, that means the message > has not been sent to the client (this event is generated when the full > message has been written into the socket). > > > coupled with the closed log > > message I have to assume it's the server closing the connection. I can't > > really see what's going on in the internals of the client as it's 3rd > party > > so all I have are the error codes. > At this point, you should first verify that the data exchanged between > your server and the client are what you expect them to be. You can use > Wireshark for that. > > On my side, it's rather difficult to tell you what can be wrong without > seing the code... > > > > > > On 3 April 2013 17:34, Emmanuel Lécharny <[email protected]> wrote: > > > >> Le 4/3/13 6:02 PM, Jennifer Cumming a écrit : > >>> Added and nothing. exceptionCaught is not getting called. Even tried > >>> putting a breakpoint in and debugging and it's not getting hit. > >> Ok, at least, that says it's not an exception... > >> > >> Isn't it possible that your client closes the connection ? > >> > >> Btw, you should set the TcpNoDelay to true, unless you are sending a lot > >> of data. > >> > >> -- > >> Regards, > >> Cordialement, > >> Emmanuel Lécharny > >> www.iktek.com > >> > >> > > > -- > Regards, > Cordialement, > Emmanuel Lécharny > www.iktek.com > >
