3

I am trying to make a client send a request to a server and receive a response whilst keeping to connection up.

If i close the socket:

//server side
outToClient.writeBytes("Message to send");
connectionSocket.close();

//client side
serverResponse = inFromServer.readLine(); 
System.out.println("FROM SERVER: " + serverResponse);

Output on client side

FROM SERVER: Message to send

And after that the connection is lost, obviously.

If i do not close the socket:

//server side
outToClient.writeBytes("Message to send");

//client side
serverResponse = inFromServer.readLine(); 
System.out.println("FROM SERVER: " + serverResponse);

No output on client side. The server never sends the message or the client never receives it.

Anyone knows a possible reason for this to be happening? The client uses a thread to receive messages and a thread to send messages. The client socket is created in the main client thread so receiver and sender threads use the current socket to communicate.

Thanks in advance.

Improve this question
4
  • 1
    What does the server do? Do it send back with a trailing newline? Because otherwise readlLine will wait for a newline (or connection closed, obviously). Commented Nov 12, 2014 at 17:21
  • Check this out: stackoverflow.com/questions/10472486/… Commented Nov 12, 2014 at 17:22
  • server uses outToClient to send the message wich is defined like this: DataOutputStream outToClient = new DataOutputStream(connectionSocket.getOutputStream()); Commented Nov 12, 2014 at 17:33
  • 1
    So, it doesn't send with a newline? That means that readLine will block until it receives a newline or a closed connection. The name of the function should give you a hint about that. Commented Nov 12, 2014 at 17:35

2 Answers 2

Reset to default
9

If the client expects to read a line, the server should write a line. A line is terminated by a \n character.

It may also be necessary to flush the stream you are using to send data to the client: 

outToClient.writeBytes("Message to send\n");
outToClient.flush();

Without seeing the code it's hard to say if the flush is required or not.

Improve this answer
Sign up to request clarification or add additional context in comments.

1 Comment

the \n solves it xD thank you so much. I knew it had to be something simple and it was driving me crazy.
3

Never, ever write any code on top of TCP without first specifying the protocol. Otherwise, when you have a problem like this, it's impossible to determine which end is at fault.

One side believes it's sending a "message". The other side does not believe the data it receives is a "message". Which is right? Well, if we had a protocol specification, we'd look at its definition of a "message" and see.

But we don't have one, so we can't say. This makes it impossible to construct a correct fix. If you change the sender to send a line, the receiver will still be broken in requiring a line. Or won't it? And you change the receiver to process messages that aren't lines, is that fixing it or breaking it?

It seems in this case that both sides are wrong. The most likely intent is that a message consists of a line of data terminated with a newline. The sender is not sending a newline and the recipient is not insisting on one either (because it accepts the data as a "message" when the connection closes). So unless the design intent really was that a "message" is a chunk of data not including a line ending terminated by either a line ending or the close of a connection, both sides are wrong.

Document the protocol. It's an essential step.

Improve this answer

Comments

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.