This question already has answers here:
Does the port change when a server accepts a TCP connection?
(3 answers)
Closed 4 years ago.
I understand the basics of how ports work. However, what I don't get is how multiple clients can simultaneously connect to say port 80. I know each client has a unique (for their machine) port. Does the server reply back from an available port to the client, and simply state the reply came from 80? How does this work?
First off, a "port" is just a number. All a "connection to a port" really represents is a packet which has that number specified in its "destination port" header field.
Now, there are two answers to your question, one for stateful protocols and one for stateless protocols.
For a stateless protocol (ie UDP), there is no problem because "connections" don't exist - multiple people can send packets to the same port, and their packets will arrive in whatever sequence. Nobody is ever in the "connected" state.
For a stateful protocol (like TCP), a connection is identified by a 4-tuple consisting of source and destination ports and source and destination IP addresses. So, if two different machines connect to the same port on a third machine, there are two distinct connections because the source IPs differ. If the same machine (or two behind NAT or otherwise sharing the same IP address) connects twice to a single remote end, the connections are differentiated by source port (which is generally a random high-numbered port).
Simply, if I connect to the same web server twice from my client, the two connections will have different source ports from my perspective and destination ports from the web server's. So there is no ambiguity, even though both connections have the same source and destination IP addresses.
Ports are a way to multiplex IP addresses so that different applications can listen on the same IP address/protocol pair. Unless an application defines its own higher-level protocol, there is no way to multiplex a port. If two connections using the same protocol simultaneously have identical source and destination IPs and identical source and destination ports, they must be the same connection.
Important:
I'm sorry to say that the response from "Borealid" is imprecise and somewhat incorrect - firstly there is no relation to statefulness or statelessness to answer this question, and most importantly the definition of the tuple for a socket is incorrect.
First remember below two rules:
Primary key of a socket: A socket is identified by {SRC-IP, SRC-PORT, DEST-IP, DEST-PORT, PROTOCOL} not by {SRC-IP, SRC-PORT, DEST-IP, DEST-PORT} - Protocol is an important part of a socket's definition.
OS Process & Socket mapping: A process can be associated with (can open/can listen to) multiple sockets which might be obvious to many readers.
Example 1: Two clients connecting to same server port means: socket1 {SRC-A, 100, DEST-X,80, TCP} and socket2{SRC-B, 100, DEST-X,80, TCP}. This means host A connects to server X's port 80 and another host B also connects to the same server X to the same port 80. Now, how the server handles these two sockets depends on if the server is single-threaded or multiple-threaded (I'll explain this later). What is important is that one server can listen to multiple sockets simultaneously.
To answer the original question of the post:
Irrespective of stateful or stateless protocols, two clients can connect to the same server port because for each client we can assign a different socket (as the client IP will definitely differ). The same client can also have two sockets connecting to the same server port - since such sockets differ by SRC-PORT. With all fairness, "Borealid" essentially mentioned the same correct answer but the reference to state-less/full was kind of unnecessary/confusing.
To answer the second part of the question on how a server knows which socket to answer. First understand that for a single server process that is listening to the same port, there could be more than one socket (maybe from the same client or from different clients). Now as long as a server knows which request is associated with which socket, it can always respond to the appropriate client using the same socket. Thus a server never needs to open another port in its own node than the original one on which the client initially tried to connect. If any server allocates different server ports after a socket is bound, then in my opinion the server is wasting its resource and it must be needing the client to connect again to the new port assigned.
A bit more for completeness:
Example 2: It's a very interesting question: "can two different processes on a server listen to the same port". If you do not consider protocol as one of the parameters defining sockets then the answer is no. This is so because we can say that in such a case, a single client trying to connect to a server port will not have any mechanism to mention which of the two listening processes the client intends to connect to. This is the same theme asserted by rule (2). However, this is the WRONG answer because 'protocol' is also a part of the socket definition. Thus two processes in the same node can listen to the same port only if they are using different protocols. For example, two unrelated clients (say one is using TCP and another is using UDP) can connect and communicate to the same server node and to the same port but they must be served by two different server processes.
Server Types - single & multiple:
When a server processes listening to a port that means multiple sockets can simultaneously connect and communicate with the same server process. If a server uses only a single child process to serve all the sockets then the server is called single-process/threaded and if the server uses many sub-processes to serve each socket by one sub-process then the server is called a multi-process/threaded server. Note that irrespective of the server's type a server can/should always use the same initial socket to respond back (no need to allocate another server port).
Suggested Books and the rest of the two volumes if you can.
A Note on Parent/Child Process (in response to query/comment of 'Ioan Alexandru Cucu')
Wherever I mentioned any concept in relation to two processes say A and B, consider that they are not related by the parent-child relationship. OS's (especially UNIX) by design allows a child process to inherit all File-descriptors (FD) from parents. Thus all the sockets (in UNIX like OS are also part of FD) that process A listening to can be listened to by many more processes A1, A2, .. as long as they are related by parent-child relation to A. But an independent process B (i.e. having no parent-child relation to A) cannot listen to the same socket. In addition, also note that this rule of disallowing two independent processes to listen to the same socket lies on an OS (or its network libraries), and by far it's obeyed by most OS's. However, one can create own OS which can very well violate this restriction.
TCP / HTTP Listening On Ports: How Can Many Users Share the Same Port
So, what happens when a server listen for incoming connections on a TCP port? For example, let's say you have a web-server on port 80. Let's assume that your computer has the public IP address of 24.14.181.229 and the person that tries to connect to you has IP address 10.1.2.3. This person can connect to you by opening a TCP socket to 24.14.181.229:80. Simple enough.
Intuitively (and wrongly), most people assume that it looks something like this:
Local Computer | Remote Computer
--------------------------------
<local_ip>:80 | <foreign_ip>:80
^^ not actually what happens, but this is the conceptual model a lot of people have in mind.
This is intuitive, because from the standpoint of the client, he has an IP address, and connects to a server at IP:PORT. Since the client connects to port 80, then his port must be 80 too? This is a sensible thing to think, but actually not what happens. If that were to be correct, we could only serve one user per foreign IP address. Once a remote computer connects, then he would hog the port 80 to port 80 connection, and no one else could connect.
Three things must be understood:
1.) On a server, a process is listening on a port. Once it gets a connection, it hands it off to another thread. The communication never hogs the listening port.
2.) Connections are uniquely identified by the OS by the following 5-tuple: (local-IP, local-port, remote-IP, remote-port, protocol). If any element in the tuple is different, then this is a completely independent connection.
3.) When a client connects to a server, it picks a random, unused high-order source port. This way, a single client can have up to ~64k connections to the server for the same destination port.
So, this is really what gets created when a client connects to a server:
Local Computer | Remote Computer | Role
-----------------------------------------------------------
0.0.0.0:80 | <none> | LISTENING
127.0.0.1:80 | 10.1.2.3:<random_port> | ESTABLISHED
Looking at What Actually Happens
First, let's use netstat to see what is happening on this computer. We will use port 500 instead of 80 (because a whole bunch of stuff is happening on port 80 as it is a common port, but functionally it does not make a difference).
netstat -atnp | grep -i ":500 "
As expected, the output is blank. Now let's start a web server:
sudo python3 -m http.server 500
Now, here is the output of running netstat again:
Proto Recv-Q Send-Q Local Address Foreign Address State
tcp 0 0 0.0.0.0:500 0.0.0.0:* LISTEN -
So now there is one process that is actively listening (State: LISTEN) on port 500. The local address is 0.0.0.0, which is code for "listening for all". An easy mistake to make is to listen on address 127.0.0.1, which will only accept connections from the current computer. So this is not a connection, this just means that a process requested to bind() to port IP, and that process is responsible for handling all connections to that port. This hints to the limitation that there can only be one process per computer listening on a port (there are ways to get around that using multiplexing, but this is a much more complicated topic). If a web-server is listening on port 80, it cannot share that port with other web-servers.
So now, let's connect a user to our machine:
quicknet -m tcp -t localhost:500 -p Test payload.
This is a simple script (https://github.com/grokit/dcore/tree/master/apps/quicknet) that opens a TCP socket, sends the payload ("Test payload." in this case), waits a few seconds and disconnects. Doing netstat again while this is happening displays the following:
Proto Recv-Q Send-Q Local Address Foreign Address State
tcp 0 0 0.0.0.0:500 0.0.0.0:* LISTEN -
tcp 0 0 192.168.1.10:500 192.168.1.13:54240 ESTABLISHED -
If you connect with another client and do netstat again, you will see the following:
Proto Recv-Q Send-Q Local Address Foreign Address State
tcp 0 0 0.0.0.0:500 0.0.0.0:* LISTEN -
tcp 0 0 192.168.1.10:500 192.168.1.13:26813 ESTABLISHED -
... that is, the client used another random port for the connection. So there is never confusion between the IP addresses.
Normally, for every connecting client the server forks a child process that communicates with the client (TCP). The parent server hands off to the child process an established socket that communicates back to the client.
When you send the data to a socket from your child server, the TCP stack in the OS creates a packet going back to the client and sets the "from port" to 80.
Multiple clients can connect to the same port (say 80) on the server because on the server side, after creating a socket and binding (setting local IP and port) listen is called on the socket which tells the OS to accept incoming connections.
When a client tries to connect to server on port 80, the accept call is invoked on the server socket. This creates a new socket for the client trying to connect and similarly new sockets will be created for subsequent clients using same port 80.
Words in italics are system calls.
Ref
http://www.scs.stanford.edu/07wi-cs244b/refs/net2.pdf
Related
I know that port numbers are used for identifying different processes running on a server, so that multiple processes can use the same networking resources. But how does it work internally?
For example, if a request to a website http://www.my-awesome-website.com:80 reaches a server, how does the server know that there is a web server running on port 80? I mean, what does the request pipeline look like between getting the request to finding out that a web server is running on port 80 and forwarding the request to the web server?
Port numbers are merely addresses for some transport-layer protocols, such as TCP and UDP, in the same way that IP addresses are for layer-3 protocols, and MAC addresses are for layer-2 protocols. Not all transport-layer protocols use ports, and each transport-layer protocol independently maintains its ports so that TCP port 80 is not the same as UDP port 80, and each can be used simultaneously by different applications.
Layer-2 addresses are only relevant to the LAN links, layer-3 addresses are only relevant host-to-host over the layer-3 network, and layer-4 addresses are relevant application-to-application.
IANA registers ports and maintains the official registry list at Service Name and Transport Protocol Port Number Registry.
From RFC 793, TRANSMISSION CONTROL PROTOCOL:
Multiplexing:
To allow for many processes within a single Host to use TCP
communication facilities simultaneously, the TCP provides a set of
addresses or ports within each host. Concatenated with the network
and host addresses from the internet communication layer, this forms
a socket. A pair of sockets uniquely identifies each connection.
That is, a socket may be simultaneously used in multiple
connections.
The binding of ports to processes is handled independently by each
Host. However, it proves useful to attach frequently used processes
(e.g., a "logger" or timesharing service) to fixed sockets which are
made known to the public. These services can then be accessed
through the known addresses. Establishing and learning the port
addresses of other processes may involve more dynamic mechanisms.
Connections:
The reliability and flow control mechanisms described above require
that TCPs initialize and maintain certain status information for
each data stream. The combination of this information, including
sockets, sequence numbers, and window sizes, is called a connection.
Each connection is uniquely specified by a pair of sockets
identifying its two sides.
When two processes wish to communicate, their TCP's must first
establish a connection (initialize the status information on each
side). When their communication is complete, the connection is
terminated or closed to free the resources for other uses.
Since connections must be established between unreliable hosts and
over the unreliable internet communication system, a handshake
mechanism with clock-based sequence numbers is used to avoid
erroneous initialization of connections.
After opening a socket(which is like an open file but used for network communications), the user of the socket may use it directly with an ephemeral port(selected by the OS), which is typical if the application is a client application.
What server processes do is to call the bind() socket API call to set a port for the socket, and then call listen() in case of a TCP socket to start listening for incoming connection requests.
Because of the bind() call the OS will know that this particular socket is the one receiving the data sent to the particular port number.
The packets sent over the network contain the source and destination IP addresses as well as the source and destination ports:
http://www.techrepublic.com/article/exploring-the-anatomy-of-a-data-packet/
So the OS has a data structure with open sockets listed by their port numbers and it will pass the received data to the correct socket's input buffer. Sent data will be marked by the port number of the sending socket.
As far as I know, a (single generic) webserver uses ports (like any other tcp/upd application) to identify users/process. Since, a port is a 16-bit unsigned integer, thus ranging from 0 to 65535. How does server act when it reaches its limit?
High level sample
The server1 answering by 8080! The client1 conect to server1 (now they are connected by random ports: (but unique) server1:5123 <--> client1:6123)
Another client2 connect to the server1 ( server1:5124 <--> client2:7123 )
So, the thing is: Is the server limited by 65535 (in pratical less than that) for a given instance?
In the simplest case a webserver consumes only one TCP port (conventionally port 80) on the server system. All of the connections to the webserver are handled through that single port. The other 65534 ports remain available for other uses.
This works because a TCP connection is identified not just by the port number on the server, but by the combination of (server IP, server TCP port, client IP, client TCP port). So the server can have a huge number of concurrent TCP connections all on its port 80, using the other three items to identify which connection the traffic belongs to. If the server only has a single IP address, and therefore the (server IP, server port) portions are identical on all connections to the webserver, the individual connections are still distinguishable by the (client IP, client port) portions of the combination.
If you run the netstat -a command on a busy Unix webserver you'll see this in action. That command will show a bunch of connections on the server's port 80, but all with different client IPs and/or ports. It'll also show that the system is still listening for new connections on port 80, at the same time as it is handling all of the existing connections on that port.
The total number of connections to the webserver might be limited by some other constraint (perhaps memory usage, perhaps some arbitrary limit in the webserver itself or in the OS kernel) or by some external constraint (perhaps connection table size in external firewalls or gateways) but it's not limited by the 16-bit TCP port range.
Also note that TCP ports are completely separate from UDP ports, so using TCP port 80 for a webserver does not prevent UDP port 80 from being used for some other purpose. And vice versa.
To my understanding by serverSocket = new ServerSocket(portNumber) we create an object which potentially can "listen" to the indicated port. By clientSocket = serverSocket.accept() we force the server socket to "listen" to its port and to "accept" a connection from any client which tries to connect to the server through the port associated with the server. When I say "client tries to connect to the server" I mean that client program executes "nameSocket = new Socket(serverIP,serverPort)".
If client is trying to connect to the server, the server "accepts" this client (i.e. creates a "client socket" associated with this client).
If a new client tries to connect to the server, the server creates another client socket (associated with the new client). But how the server knows if it is a "new" client or an "old" one which has already its socket? Or, in other words, how the clients are identified? By their IP? By their IP and port? By some "signatures"?
What happens if an "old" client tries to use Socket(serverIP,serverIP) again? Will server create the second socket associated with this client?
The server listens on an address and port. For example, your server's IP address is 10.0.0.1, and it is listening on port 8000.
Your client IP address is 10.0.0.2, and the client "connects" to the server at 10.0.0.1 port 8000. In the TCP connect, you are giving the port of the server that you want to connect to. Your client will actually get its own port number, but you don't control this, and it will be different on each connection. The client chooses the server port that it wants to connect to and not the client port that it is connecting from.
For example, on the first connection, your client may get client-side port 12345. It is connecting from 10.0.0.2 port 12345 to the server 10.0.0.1 port 8000. Your server can see what port the client is connecting from by calling getpeername on its side of the connection.
When the client connects a second time, the port number is going to be different, say port 12377. The server can see this by calling getpeername on the second connection -- it will see a different port number on the client side. (getpeername also shows the client's IP address.)
Also, each time you call accept on the server, you are getting a new socket. You still have the original socket listening, and on each accept you get a new socket. Call getpeername on the accepted socket to see which client port the connection is coming from. If two clients connect to your server, you now have three sockets -- the original listening socket, and the sockets of each of the two clients.
You can have many clients connected to the same server port 8000 at the same time. And, many clients can be connected from the same client port (e.g. port 12345), only not from the same IP address. From the same client IP address, e.g. 10.0.0.2, each client connection to the server port 8000 will be from a unique client port, e.g. 12345, 12377, etc. You can tell the clients apart by their combination of IP address and port.
The same client can also have multiple connections to the server at the same time, e.g. one connection from client port 12345 and another from 12377 at the same time. By client I mean the originating IP address, and not a particular software object. You'll just see two active connections having the same client IP address.
Also, eventually over time, the combination of client-address and client-port can be reused. That is, eventually, you may see a new client come in from 10.0.0.2 port 12345, long after the first client at 10.0.0.2 port 12345 has disconnected.
Every TCP connection has as identifier the quadruple (src port, src address, dest port, dest address).
Whenever your server accepts a new client, a new Socket is created and it's indipendent from every other socket created so far. The identification of clients is not implictly handled somehow..
You don't have to think sockets as associated to "clients", they are associated with an ip and a port, but there is not direct correlation between these two.
If the same client tries to open another socket by creating a new one you'll have two unrelated sockets (because ports will be different for sure). This because the client cannot use the same port to open the new connection so the quadruple will be different, same client ip, same server ip, same server port but different client port.
EDIT for your questions:
clients don't specify a port because it's randomly choosen from the free ones (> 1024 if I'm not wrong) from the underlying operating system
a connection cannot be opened from a client using the same port, the operating system won't let you do that (actually you don't specify any port at all) and in any case it would tell you that port is already bound to a socket so this issue cannot happen.
whenever the server receives a new connection request it's is considered new, because also if ip is the same port will be different for sure (in case of old packet resend or similar caveats I think that the request will be discarded)
By the way all these situations are clearly explained in TCP RFC here.
I think the question here is why do you care if the client is new or old. What is new and old?
For example, a web browser could connect to a web server to request a web page. This will create a connection so serverSocket.accept() will return a new Socket. Then the connection is closed by the web browser.
Afer a couple of minutes, the end used click on a link in the web page and the browser request a new page to the server. This will create a connection so serverSocket.accept() will return a new Socket.
Now, the web server do not care if this is a new or old client. It just need to server the requested page. If the server do care if the "client" already requested a page in the past, it should do so using some information in the protocol used on the socket. Check out http://en.wikipedia.org/wiki/OSI_model
In this case, the ServerSocket and Socket ack on the transport level. The question "does this client already requested a page on the server" should be answered by information on the session or even application layer.
In the web browser/server example, the http protocol (which is an application) protocol hold information about who is this browser in the parameters of the request (the browser transmit cookie informations with every request). The http server can then set/read cookie information to known if the browser connected before and eventually maintain a server side session for that browser.
So back to your question: why do you care if it's a new or old client?
A socket is identified by:
(Local IP,Local Port, Remote IP,
Remote Port,IP Protocol(UDP/TCP/SCTP/etc.)
And that's the information the OS uses to map the packets/data to the right handle/file descriptor of your program. For some kinds of sockets,(e.g. an non-connected UDP socket)the remote port/remote IP might be wildcards.
By definition, this is not a Java related question, but about networking in general, since Sockets and SeverSockets apply to any networking-enabled programming language.
A Socket is bounded to a local-port. The client will open a connection to the server (by the Operating System/drivers/adapters/hardware/line/.../line/hardware/adapters/drivers/Server OS). This "connection" is done by a protocol, called the IP (Internet Protocol) when you are connected to the Internet. When you use "Sockets", it will use another protocol, which is the TCP/IP-protocol.
The Internet Protocol will identify nodes on a network by two things: their IP-address and their port. The TCP/IP-protocol will send messages using the IP, and making sure messages are correctly received.
Now; to answer your question: it all depends! It depends on your drivers, your adapters, your hardware, your line. When you connect to your localhost machine, you will not get further than the adapter. The hardware isn't necessairy, since no data is actually sent over the line. (Though often you need hardware before you can have an adapter.)
By definition, the Internet Protocol defines a connection as pair of nodes (thus four things: two IP-adresses and two ports). Also, the Internet Protocol defines that one node can only use one port at a time to initiate a connection with another node (note: this only applies for the client, not the server).
To answer your second question: if there are two Sockets: the "new" and the "old". Since, by the Internet Protocol, a connection is a pair of nodes, and nodes can only use one port at a time for a connection, the ports of "new" and "old" must be different. And because this is different, the "new" client can be discriminated from the "old", since the port-number is differently.
From what I understand, each HTTP request uses its own TCP connection (please correct me if i'm wrong). So, let's say that there are two current connections to the same server. For example, client side javascript code triggering a couple of AJAX POST requests using the XMLHttpRequest object, one right after the other, before getting the response to the first one. So we're talking about two connections to the same server, each waiting for a response in order to route it to each separate callback function.
Now here's the thing that I don't understand: The TCP packet includes source and destination ip and port, but won't both of these connections have the same src and dest ip addresses, and port 80? How can the packets be differentiated and routed to appropriately? Does it have anything to do with the packet sequence number which is different for each connection?
When your browser creates a new connection to the HTTP server, it uses a different source port.
For example, say your browser creates two connections to a server and that your IP address is 60.12.34.56. The first connection might originate from source port 60123 and the second from 60127. This is embedded in the TCP header of each packet sent to the server. When the server replies to each connection, it uses the appropriate port (e.g. 60123 or 60127) so that the packet makes it back to the right spot.
One of the best ways to learn about this is to download Wireshark and just observe traffic on your own network. It will show you this and much more.
Additionally, this gives insight into how Network Address Translation (NAT) works on a router. You can have many computers share the same IP address and the router will rewrite the request to use a different port so that two computers can simultaneously connect to places like AOL Instant Messenger.
They're differentiated by the source port.
The main reason for each HTTP request to not generate a separate TCP connection is called keepalives, incidentally.
A socket, in packet network communications, is considered to be the combination of 4 elements: server IP, server port, client IP, client port. The second one is usually fixed in a protocol, e.g. http usually listen in port 80, but the client port is a random number usually in the range 1024-65535. This is because the operating system could use those ports for known server protocols (e.g. 21 for FTP, 22 for SSH, etc.). The same network device can not use the same client port to open two different connections even to different servers and if two different clients use the same port, the server can tell them apart by their IP addresses. If a port is being used in a system either to listen for connection or to establish a connection, it can not be used for anything else. That's how the operating system can dispatch packets to the correct process once received by the network card.
The socket API is the de-facto standard for TCP/IP and UDP/IP communications (that is, networking code as we know it). However, one of its core functions, accept() is a bit magical.
To borrow a semi-formal definition:
accept() is used on the server side.
It accepts a received incoming attempt
to create a new TCP connection from
the remote client, and creates a new
socket associated with the socket
address pair of this connection.
In other words, accept returns a new socket through which the server can communicate with the newly connected client. The old socket (on which accept was called) stays open, on the same port, listening for new connections.
How does accept work? How is it implemented? There's a lot of confusion on this topic. Many people claim accept opens a new port and you communicate with the client through it. But this obviously isn't true, as no new port is opened. You actually can communicate through the same port with different clients, but how? When several threads call recv on the same port, how does the data know where to go?
I guess it's something along the lines of the client's address being associated with a socket descriptor, and whenever data comes through recv it's routed to the correct socket, but I'm not sure.
It'd be great to get a thorough explanation of the inner-workings of this mechanism.
Your confusion lies in thinking that a socket is identified by Server IP : Server Port. When in actuality, sockets are uniquely identified by a quartet of information:
Client IP : Client Port and Server IP : Server Port
So while the Server IP and Server Port are constant in all accepted connections, the client side information is what allows it to keep track of where everything is going.
Example to clarify things:
Say we have a server at 192.168.1.1:80 and two clients, 10.0.0.1 and 10.0.0.2.
10.0.0.1 opens a connection on local port 1234 and connects to the server. Now the server has one socket identified as follows:
10.0.0.1:1234 - 192.168.1.1:80
Now 10.0.0.2 opens a connection on local port 5678 and connects to the server. Now the server has two sockets identified as follows:
10.0.0.1:1234 - 192.168.1.1:80
10.0.0.2:5678 - 192.168.1.1:80
Just to add to the answer given by user "17 of 26"
The socket actually consists of 5 tuple - (source ip, source port, destination ip, destination port, protocol). Here the protocol could TCP or UDP or any transport layer protocol. This protocol is identified in the packet from the 'protocol' field in the IP datagram.
Thus it is possible to have to different applications on the server communicating to to the same client on exactly the same 4-tuples but different in protocol field. For example
Apache at server side talking on (server1.com:880-client1:1234 on TCP)
and
World of Warcraft talking on (server1.com:880-client1:1234 on UDP)
Both the client and server will handle this as protocol field in the IP packet in both cases is different even if all the other 4 fields are same.
What confused me when I was learning this, was that the terms socket and port suggest that they are something physical, when in fact they're just data structures the kernel uses to abstract the details of networking.
As such, the data structures are implemented to be able to distinguish connections from different clients. As to how they're implemented, the answer is either a.) it doesn't matter, the purpose of the sockets API is precisely that the implementation shouldn't matter or b.) just have a look. Apart from the highly recommended Stevens books providing a detailed description of one implementation, check out the source in Linux or Solaris or one of the BSD's.
As the other guy said, a socket is uniquely identified by a 4-tuple (Client IP, Client Port, Server IP, Server Port).
The server process running on the Server IP maintains a database (meaning I don't care what kind of table/list/tree/array/magic data structure it uses) of active sockets and listens on the Server Port. When it receives a message (via the server's TCP/IP stack), it checks the Client IP and Port against the database. If the Client IP and Client Port are found in a database entry, the message is handed off to an existing handler, else a new database entry is created and a new handler spawned to handle that socket.
In the early days of the ARPAnet, certain protocols (FTP for one) would listen to a specified port for connection requests, and reply with a handoff port. Further communications for that connection would go over the handoff port. This was done to improve per-packet performance: computers were several orders of magnitude slower in those days.