socket disconnection notification method - c#

just searched for a posibble solution to indetify when the client disconnecets.
i found this:
public bool IsConnected( Socket s)
{
try
{
return !(s.Poll(1, SelectMode.SelectRead) &&s.Available == 0);
}
catch (SocketException) { return false; }
}
im using a while loop in my main with thread.sleep(500) and running the Isconnectedmthod it works allright when i run it through the visual studio and when i click stop debugging it actually notify me in the server side program but when i just go to the exe in the bin directory and launch it-it's Indeed notify me for a connection but when i close the program (manually from the 'x' button) or through the task manager theIsConnected method apparently return still true.....
im using a simple tcp connection
client = new TcpClient();
client.Connect("10.0.0.2", 10);
server:
Socket s = tcpClient.Client;
while(true)
{
if (!IsConnected(s))
MessageBox.Show("disconnected");
}
(it's running on a thread btw).
any suggestion guys?
i even tried to close the connection when the client closes:
private void Form1_FormClosing(object sender, FormClosingEventArgs e)
{
client.Close();
s.Close();
Environment.Exit(0);
}
dont know what to do

What you are asking for is not possible. TCP will not report an error on the connection unless an attempt is made to send on the connection. If all your program ever does is receive, it will never notice that the connection no longer exists.
There are some platform-dependent exceptions to this rule, but none involving the simple disappearance of the remote endpoint.
The correct way for a client to disconnect is for it to gracefully close the connection with a "shutdown" operation. In .NET, this means the client code calls Socket.Shutdown(SocketShutdown.Send). The client must then continue to receive until the server calls Socket.Shutdown(SocketShutdown.Both). Note that the shutdown "reason" is generally "send" for the endpoint initiating the closure, and "both" for the endpoint acknowledging and completing the closure.
Each endpoint will detect that the other endpoint has shutdown its end by the completion of a receive operation with 0 as the byte count return value for that operation. Neither endpoint should actually close the socket (i.e. call Socket.Close()) until this two-way graceful closure has completed. I.e. each endpoint has both called Socket.Shutdown() and seen a zero-byte receive operation completion.
The above is how graceful closure works, and it should be the norm for server/client interactions. Of course, things do break. A client could crash, the network might be disconnected, etc. Typically, the right thing to do is to delay recognition of such problems as long as possible; for example, as long as the server and client have no need to actually communicate, then a temporary network outage should not cause an error. Forcing one is pointless in that case.
In other words, don't add code to try to detect a connection failure. For maximum reliability, let the network try to recover on its own.
In some less-common cases, it is desirable to detect connection failures earlier. In these cases, you can enable "keep alive" on the socket (to force data to be sent over the connection, thus detecting interruptions in the connection…see SocketOptionName.KeepAlive) or implement some timeout mechanism (to force the connection to fail if no data is sent after some period of time). I would generally recommend against the use of this kind of technique, but it's a valid approach in some cases.

Related

When a TCP connection is terminated, how is either side informed?

Let's say we have a basic TCP server with a .NET TcpListener, with a basic TCP client using a .NET TcpClient.
What types of connection terminations are there, and how are they supposed to be checked for and handled?
Client
A. Client gracefully terminates the connection. The server is notified.
B. Client physically disconnects from the network. How does the server know?
C. Client program shuts down without a graceful disconnect. How does the server know?
Server
A. Server gracefully terminates the connection. The client is notified.
B. Server physically disconnects from the network. How does the client know?
C. Server program shuts down without a graceful disconnect. How does the client know?
Cases A and C are communicated by a TCP packet with the FIN flag set in the header. It is sent by the TCP/IP stack in the OS, so it doesn't matter if the application exited abnormally. The subcase of C where the OS also failed will act like B instead.
Case B, when you've lost the ability to communicate, is more complicated. If the failure is local (e.g. disassociation from a WiFi access point), then the local end of the connection will find out immediately about a change in network status, and can infer that the connection is broken (but if not cleaned up, a connection can survive a short-term outage).
If the connections are actively transmitting data, the acknowledgements will timeout and result in retransmission attempts. A limit may be placed on retransmission attempts, resulting in an error.
If there is no traffic, then it is possible for loss of connection to go undetected for a very long time (multiple days). For this reason, TCP connections often are configured to send heartbeat packets, which must be acknowledged, and detect failure of retransmission attempts in the same manner as normal data.
The short answer is: neither knows instantly when the other disconnects. UNTIL of course something lets you know.TCP isn't a physical connection therefore it relies on signals of some sort to determine the state. Usually either by checking the connection or receiving a graceful disconnect message from the other side. There are various ways to check for connections (polling, timeouts, catching socket exceptions, etc). It really depends on framework you're using and what your needs are.
If you have the resources to poll, then you can poll ever x seconds and check the state. Etc.
Check this extension method found here.
static class SocketExtensions
{
public static bool IsConnected(this Socket socket)
{
try
{
return !(socket.Poll(1, SelectMode.SelectRead) && socket.Available == 0);
}
catch (SocketException) { return false; }
}
}
Basically, socket.Poll() returns true if the connection is open and there is data, or it will return true if there is no connection. socket.Available will return the number of bytes being sent. If it's 0, && .Poll() was true, then the connection is effectively closed. It gets confusing if you try to follow the true/falses, but this will return with pretty good accuracy.

socket doesn't close when thread done

I have a problem with sockets. This:
When client-thread ends, server trying to read, and its freezes, because socket is not closed. Thread dont close it, when its over. Its problem exist, if i using thread, but if i using two independents projects, i have no problem (exception throws, and i can catch it).
I cant use timeout, and i must correct continue server-work, when client dont close socket.
Sorry for my bad eng.
As far as I know, there is no way for TCP server (listener) to find out whether data from client are not coming because it has died/quit or is just inactive. This is not .NET's defficiency, it is how TCP works. The way I deal with it is:
1. Create a timer in my client that periodically sends signal "I am alive" to the server. For example, I just send 1 unusual ASCII character '∩' (code 239).
2. In TCP listener: use NetworkStream.Read(...) method that allows to specify timeout. If timeout expires, the server disposes the old NetworkStream instance and creates new one on the same TCP port. If the server receives "I am alive" signal from client, it keeps listening.
By the way, the property TcpClient.Connected is useless for detecting on server side whether client still uses the socket. The method only returns true if last Read action returned something. So, if client is alive and just silent, the TcpClient.Connected becomes false.
Close client when you want the connection to be closed (at the end of Client).
Better yet, use using for all disposable resources such as both clients and the listener.

What are the possible reasons of SocketError.ConnectionReset in TCP Socket

I have a TCP socket based client server system.
Everything works fine but when network is disconnected form client end and reconnect it again i get automatically SocketError.ConnectionReset send form client and regarding this command the socket is closed in the server side. this is also fine.
but when i look in to the client side it shows the socket is still connected with server. (regarding socket is still connected with server [It does not happen every time], sometime it shows disconnected and some times shows connected)
Does it make sense that "server get a SocketError.ConnectionReset from
client end but client is still connected"?
So i want to know what is the possible reasons of SocketError.ConnectionReset and how to handle such type of problem i have mentioned?
Again i say, Everything is working fine in normal environment (e.g if i exit the client it is disconnected the socket same for the server)
Thanks in advance.
EDIT:
Here is the code in the client side. actually it's a timer that tick every 3 second through programs lifetime and check if Socket is connected or not if its disconnected then it tries to reconnect again through a new socket instance
private void timerSocket_Tick(object sender, EventArgs e)
{
try
{
if (sck == null || !sck.Connected)
{
ConnectToServer();
}
}
catch (Exception ex)
{
RPLog.WriteDebugLog("Exception occcured at: "+ System.Reflection.MethodBase.GetCurrentMethod().ToString()+"Message: "+ex.Message);
}
}
In normal situation (without network disconnect/reconnect) if TCP server get a
SocketError.ConnectionReset form any client, in the client side i see
clients socket is disconnected and it tries to reconnect it again
through the code shown. but when situation happen explained earlier,
server gets a SocketError.ConnectionReset but client shows it still
connected. though the TCP server shows the reset command is send form the exact client
side.
There are several causes but the most common is that you have written to a connection that has already been closed but he other end. In other words, an application protocol error. When it happens you have no choice but to close the socket, it is dead. However you can fix the underlying cause.
When discussing a TCP/IP issue like this, you must mention the network details between the client and the server.
When one side says the connection is reset, it simply means that on the wire a RST packet appears. But to know who sends the RST packet and why, you must utilize network packet captures (by using Wireshark and any other similar tools),
https://en.wikipedia.org/wiki/Transmission_Control_Protocol
You won't easily find out the cause at .NET Framework level.
The problem with using Socket.Connected as you are is that it only gives you the connected state as at the last Send or Receive operation. i.e. It will not tell you that the socket has disconnected unless you first try to send some data to it or receive data from it.
From MSDN description of the Socket.Connected property:
Gets a value that indicates whether a Socket is connected to a remote host as of the last Send or Receive operation.
So in your example, if the socket was functioning correctly when you last sent or received any data from it, the timerSocket_Tick() method would never call ConnectToServer(), even if the socket was now not connected.
how to handle such type of problem i have mentioned?
Close the socket and initiate a new connection.

Closing System.Net.Sockets.TcpClient kills the connection for other TCPClients at the same IP Address

Just to be clear, all of the TCPClients I'm referring to here are not instances of my own class, they are all instances of System.Net.Sockets.TcpClient from Mono's implementation of .NET 4.0.
I have a server that is listening for client connections, as servers do. Whenever it gets a new client it creates a new TCPClient to handle the connection on a new thread. I'm keeping track of all the connections and threads with a dictionary. If the client disconnects, it sends a disconnect message to the server, the TCPClient is closed, the dictionary entry is removed and the thread dies a natural death. No fuss, no muss. The server can handle multiple clients with no problem.
However, I'm simulating what happens if the client gets disconnected, doesn't have a chance to send a disconnect message, then reconnects. I'm detecting whether a client has reconnected with a username system (it'll be more secure when I'm done testing). If I just make a new TCPClient and leave the old one running, the system works just fine, but then I have a bunch of useless threads lying around taking up space and doing nothing. Slackers.
So I try to close the TCPClient associated with the old connection. When I do that, the new TCPClient also dies and the client program throws this error:
E/mono (12944): Unhandled Exception: System.IO.IOException: Write failure ---> System.Net.Sockets.SocketException: The socket has been shut down
And the server throws this error:
Unable to write data to the transport connection: An established connection was aborted by the software in your host machine.
Cannot read from a closed TextReader.
So closing the old TCPClient with a remote endpoint of say: 192.168.1.10:50001
Also breaks the new TCPClient with a remote endpoint of say:192.168.1.10:50002
So the two TCPClient objects have the same remote endpoint IP address, but different remote endpoint ports. But closing the one seems to stop the other from working. I want to be able to close the old TCPClient to do my cleanup, without closing the new TCPClient.
I suspect this is something to do with how TCPClient works with sockets at a low level, but not having any real understanding of that, I'm not in a position to fix it.
I had a similar issue on my socket server. I used a simple List instead of a dictionary to hold all of my current connections. In a continuous while loop that listens for new streams, I have a try / catch and in the catch block it kills the client if it has disconnected.
Something like this on the sever.cs:
public static void CloseClient(SocketClient whichClient)
{
ClientList.Remove(whichClient);
whichClient.Client.Close();
// dispose of the client object
whichClient.Dispose();
whichClient = null;
}
and then a simple dispose method on the client:
public void Dispose()
{
System.GC.SuppressFinalize(this);
}
EDIT: this paste is the OPs resolution which he or she found on their own with help from my code.
So to clarify, the situation is that I have two TCPClient objects TCPClientA and TCPClientB with different remote endpoints ports, but the same IP:
TCPClientA.Client.RemoteEndPoint.ToString();
returns: 192.168.1.10:50001
TCPClientB.Client.RemoteEndPoint.ToString();
returns: 192.168.1.10:50002
TCPClientA needs to be cleaned up because it's no longer useful, so I call
TCPClientA.Close();
But this closes the socket for the client at the other end of TCPClientB, for some reason. However, writing
TCPClientA.Client.Close();
TCPClientA.Close();
Successfully closes TCPClientA without interfering with TCPClientB. So I've fixed the problem, but I don't understand why it works that way.
Looks like you have found a solution but just so you are aware there are many similar pitfalls when writing client/server applications in .net. There is an open source network library (which is fully supported in mono) where these problems have already been solved, networkComms.net. A basic sample is here.
Disclaimer: This is a commercial product and I am the founder.
This is clearly an error in your code. Merely closing one inbound connection cannot possibly close another one. Clearly something else is happening elsewhere in your code.

C# Client-Server application problem

I run my application on a network and in some cases the client lost connection to the server. After this time, when I wanted to send a message to the server I receive the following error: Operation not allowed on non-connected sockets (something like this).
I thought to create an event for object type TcpClient and when tcp_obj.Connected = false to call a function to discontinue execution of the current code. How could I do this?
Or giving me other suggestios.
Thanks.
I know at least from socket programming in Java that when a client loses connection to the server, the server does not and can not know about it. You need a heartbeat of some sort to detect the early disconnection.
We often use a heartbeat in our client/server applications to detect early disconnections and log them on the server. This way the server can close the associated socket and release the connection back to the pool.
Simply send a command to the client periodically and wait for a response. If no response is garnered within a timeout assume disconnect and close streams.
I would simply first check your connection object to ensure you are connected, prior to attempting to send the message. Also make sure that you are putting your send-logic inside of a try-catch, so that if you do happen to get disconnected mid transmission, you'll be able to resume without blowing your application apart.
Psuedo-Code:
private void SendMessage(string message, Socket socket)
{
if(socket.connectionState = States.Connected)
{
try{
// Attempt to Send
}
catch(SocketException Ex)
{
// Disconenct, Additional Cleanup Etc.
}
}
}
If you are in C#, prior to your connection state changing, you will have a socket disconnected event fire, prior to your connection state changing. Make sure you tie this event up as soon as your socket connects.
Can we know why you use TCP sockets? Is for calling a tcp device o server code?
I recommend you if is for calling a .net server app use Windows Communication Foudation. It is simple to expose services by net.tcp, http, etc.
Regards,
Actually this is a very old problem,
If I understand your question correctly you need a way to know whether you're application is still connected to the server or vice versa.
If so then a workaround is to have a UDP connection just to check the connectivity (overhead I know, but its much better then polling on Connected state), you could check just before you send you're data.
Since UDP is not Connection oriented you don't need to be connected when you send the data

Categories

Resources