C# Socket.Receive() data is out of order & Truncated at the beginning & at End

Question

I and trying to make a TCPServer & TCPClient simultaneously in C# (DotNET 4.7.2) on Windows. From Client I send data by calling Socket.Send() consecutively 4 times, & these 4 Consecutive Sends together are in loop. On the server side I receive Data by looping through until Data available & within loop I print each byte array on the console.

now comes my Doubt:-

1. some times data comes out of order (not in order in which I called Socket.Send() at client ).

2. data in the beginning & at the end is truncated.Now people might say that first i need to collect all the bytes & after the receive loop is finished then I should concatenate it and print,Correct,I got it. But The first initial bytes itself are missing then what ?? First byte should always have been same no matter whatever the way i print it,but the inital bytes itself are truncated

3. how does SocketFlags affect this, specially partial flag ?

4. does routing affect this ?

5. Difference between calling send consecutively & sending all at once ? does TCP handles both conditions differently at protocol level or what way it handles consecutive sends at low protocol level ?

6. is it possible that that 2nd Send() method call data arrived before the 1st send() method call, in consecutive send method calling ?

I am new to socket Programming.

I am attaching the output screenshots also .In the screenshots look at the text marked by yellow.

console ScreenShot1

console Screenshot2

console screenshot3

my code is here :-

This is client side code

    static void Main(string[] args)
    {
        Console.WriteLine("press any key to start");
        Console.ReadLine();

        byte[] data = Encoding.ASCII.GetBytes("@123456789@");
        byte[] data2 = Encoding.ASCII.GetBytes("&.RAJA.BHOJ.&");
        byte[] data3 = Encoding.ASCII.GetBytes("#abcdefghi#");
        byte[] data4 = Encoding.ASCII.GetBytes("/ABCDEFGHI/");

        Socket s = new Socket(AddressFamily.InterNetwork,SocketType.Stream,ProtocolType.Tcp);
        s.Connect(ipEnd);

    mark1:
        
        int x = s.Send(data);
            x = s.Send(data2);
            x = s.Send(data3);
            x = s.Send(data4);            

        goto mark1;
   }

This is Sever Side Code

    static void Main(string[] args)
    {
        byte[] Bytes = new byte[50];
        Socket listener = new Socket(AddressFamily.InterNetwork, SocketType.Stream, protocolType.Tcp);
        listener.Bind(ipEnd);
        listener.Listen(100);

        Socket handler = listener.Accept();
    mark1:
        
        while(handler.Available != 0)
        {
            int byteCount = handler.Receive(Bytes);
            string text = System.Text.Encoding.ASCII.GetString(Bytes);
            Console.WriteLine("received data is :: " + text);
        }     
        
        goto mark1;
    }

Answer 1

The data should never be out of order, however: checking .Available is usually wrong - fortunately it doesn't really matter in this case, because your goto makes it basically a hot loop. In reality, you should usually loop until byteCount is non-positive (unless you're doing zero-length reads, which you aren't). However: you aren't applying any framing. In TCP, reads don't match 1:1 with writes; it is a stream protocol that just offers "the right bytes in the right order, or a dead socket" - it doesn't do anything about messages. As such, it is up to the protocol to define some mechanism for splitting messages. For a text protocol, a line-ending of some variety is common, or a sentinel marker byte such as a non-printing reserved character like nul (byte zero). Your code would be responsible for finding these markers and splitting the input data correctly.