Develop Reference

c# socket server - deploying, starting, stopping

I'm a little confused when it comes to socket servers in terms of deploying it online.
Running locally is fine as most tutorials get you to make a server application and a client application which I can execute. Done all that for awhile now and I'm happy with it but now I want to try it using a web host.
How would I deploy the socket server to my web host and then run the server? do I just upload the program to the server and run www.mywebpage.com/mysocketserver (assuming the program is called mysocketserver.exe)
may sound like a stupid question but I'm having one of those brain dead moments.
[Edit]
Great answers guys thank you. Shame I can only mark one as the answer.

Most hosting services are website related: you upload your website to their server and the host takes care of provisioning the application.
However you haven't written a website that is served by a web server, but a socket server, which is more akin to the web server itself that runs a website.
As a result you will need a host that will allow you to install and run applications, such as a Virtual Private Server service, rent a physical server or use a cloud service such as Amazon EC2.
You usually Remote Desktop or SSH into the server you are renting (or own) to upload and start the application either as a normal application, a Service or a Daemon.
Once you have installed your socket server on the host server and started the application running you should be able to contact your service using the IP address for your server and the port that you are running your socket server on.
For example, if your ip address is 10.0.0.1 and you've written your socket server to listen on port 1234, you should be able to contact your service at address 10.0.0.1:1234.
Take into account firewalls allowing access to that port.
You will also then be able to use a DNS service such as DynDns to assign a domain name to that ip address.

You can't do that unless your socket server is using HTTP as a protocol.
You are using a port for your server, right? The ports are used to identify which application to talk with.
When you browse the web using "http://something" you really say let me get the stuff which can be found on the IP 1.2.3.4 (DNS lookup) using port 80 (which is registered for HTTP). You don't have to specify the port in the browser since all browsers know that HTTP uses port 80.
So what you are really should do is to put the socket server on your host and tell your customers/users that they can connect to your socket server at port XXX on host "www.mywebpage.com".
If you've built a client you'll just hard code the port in it or specify the default port automatically.

The problem in the Internet is the routing/name translation, so that mywebpage gets translated into the correct IP address.
Either your webserver needs a fixed public IP address or you need to use a service like dynds which will dynamically map your changing IP address to the static name.

Add a NAT UPnP mapping on a complex network using C#

I am creating an application that listens for connections (server) from a different application (client) via a tcp connection. So if both application where to be on the same network it will be easy to establish the connection. so I am trying to add a method to the server so that users do not have to open ports on their routers in order to make the app work when using it over the internet. so let me show you a few different scenarios:
1)---------------------------------------------------SCENARIO_1-------------------------------------------------------------
the server is a computer in a house. That computer is connected to a router and that router is connected to the internet. The client is a computer on a office that has access to the internet as well.
for this scenario to work, before I used to open the ports on the house router and forward them to the server computer. on the client I just had to supply the correct IP address and same port in order to establish the connection.
now I avoided opening the ports on the house router with a really cool technique:
first I add this reference:
then I tell the router which ports I want to be forwarded to my computer (the server):
NATUPNPLib.UPnPNATClass upnpnat;
NATUPNPLib.IStaticPortMappingCollection mappings;
public ServerExample()
{
InitializeComponent();
// server local IP address
mappings.Add(1300, "TCP", 1300, "192.168.0.120", true, "my server");
//.... etc
..
when doing this I am able to connect from my office computer by providing port 1300, the WAN ip address of the server computer which is a different one. (that address can be found at: whatismyipaddress.com) THE COOL THING IS THAT I DO NOT HAVE TO DO ANY CONFIGURATION ON THE ROUTER!!!
2)-----------------------------------------------SCENARIO_2----------------------------------------------
know here comes my problem. The server happens to be on a office. that server is connected to router A, router A is connected to router B and router B is connected to the internet. In other this example is like the last example on the server with an extra router in between. and the client is somewhere else. we do not care how the network setup is on the client computer as long as it has internet access.
the way I used to solve this was by forwarding the packages from router x to the server computer and also port forwarding the traffic from port x of router B to the ip address of router A. when setting up that configuration I was able to establish a connection with the client when providing the WAN ip address of the server. (the ip address that is shown at: whatismyipaddress.com when retrieved from the network of the office).
so it will be nice if I can avoid doing all this configuration on the routers and do something similar to:
public delegate void SomeDelegate(string parameters);
NATUPNPLib.UPnPNATClass upnpnat;
NATUPNPLib.IStaticPortMappingCollection mappings;
public ServerExample()
{
InitializeComponent();
// server local IP address
mappings.Add(1300, "TCP", 1300, "192.168.150.141", true, "my server");
mappings.Add(1300, "TCP", 1300, "another ip", true, "router's ip");
so I have been playing around with that and I have not been able to make it work. also it will be nice if I could find out if the server connection is like scenario 1 with c# or scenario 2 or maybe a different scenario so that the users setting up the server do not have to specify all that info. for example I am sure that limewire does something similar.
Edit:
here is scenario 1 illustrated:
and here is scenario 2 illustrated

The feature that you are using is called Universal Plug and Play (UPnP) which is basically a way for a device to tell another device how to communicate with it. What you are doing is telling a computer to tell the router what ports you want forwarded to it. Unfortunately UPnP is a non-routable protocol which means that it can't jump across routers. So the short answer is that doing what you are trying will not work with multiple routers without manually configuring the ones "farthest" away from the client.
It might be possible to do something with Internet Gateway Device protocol (IDG) but you'd have to research that more and the library you are targeting doesn't support that. Post this question to Server Fault and you might get a better answer. (Don't post your code, just explain the basics of your problem, that you're trying to have a machine register itself with a router using UPnP which is working but you have another router in front of that that you want to automatically forward, too.)

Connect clients

I am trying to connect 2 or more computers running a c# application. All TCP examples I find only show how to connect them if they are running on the same network.
My question is how to connect them over the internet?
I want to create a service like MSN but with direct connection (like a torrent software does),
any help?
regards!

Connecting 2 computers over the internet should, mostly, be the same as connecting them on a LAN. Your problems will mainly be with the firewall. In your C# code, you will have to assign an IP address & port for both computers. You should ensure that your firewall will allow this port through without blocking (beware of security vulnerabilities when opening new ports).
If your computers don't have public IPs, you will also have to set up some NAT on your router. This will map a port on your router + your public IP address to your local computers IP address + port. The other computer will then configure this public IP address + port in their configuration. There is too many configurable things that have the potential to screw up to list in a single post. I'd suggest learning some basics on networks.
However, having said all that, in terms of C# code you shouldn't have to change a thing to get the program to work over the internet if it works over a LAN.

Can lidgren-network library connect directly peer to peer given both IP addresses

I am using lidgen-network library 3 to try and create a peer to peer connection. I am new to network programming which is why i'm using this library.
Using the DiscoverKnownPeer() function i have been able to connect within my network. What i have read online is that i need to do a NAT punchthrough which requires a running server that is outside the router. I am hoping for a solution that doesnt require this extra server.
If i can store both ip addesses on a web server, and get create a web service to send the ip's to each respective client through XML, would I be able to create a peer to peer connection?
Thank you in advance.

No, you won't be able to establish a connection through NAT without a rendezvous server
NAT stands in your way because even if you know the IP address, that only gets you as far as the router. The router receives a request coming from the internet, and doesn't know which internal IP to send it to, so it throws it away.
NAT punchthrough works by making both peers send outgoing request to each other, the router then "knows" that packets directed at its public IP on the given port should go to the given computer.
So, you need to connect both computers to an external server not behind NAT, and then use that external server to coordinate connecting the two NAT obscured peers together (this is the peer introduction thing that Lidgren has built in).

I know this is old but...
You didn't specify that using the DiscoverKnownPeer() was necessary in your setup. If it isn't then you can setup peer to peer by appplying port forwarding rules in your router.

how can i send data over internet to any specific computer in subnet

I want to know how I can send or recieve data over internet to/from a computer in subnet
(this is specially in context to PPP users bcoz getting static IP is not so much in practice).
I actually want to create an application which can transfer file between 2 specific computer in WAN.
so what are things I need to know about to do the same..(ex. PRESENT IP or MAC ADDRESS etc..)
PROGRAMATICAL EXPLANATION ALTHOUGH PREFFERED,BUT IS NOT NECCESARY...

FTP?

There is a vast torrent of useful results in google, I seriously suggest to google before you ask here.
For instance, have a look at the top result: http://www.devarticles.com/c/a/C-Sharp/Network-Programming-in-C-sharp/
Apart from that, FTP, as suggested by Colin, may be what you're looking for. If you're new to using FTP in C# have a look at http://www.google.com/search?q=c%23+ftp

Put the information on a public IP server, so both computers keep polling if there is new data and send / download that data as needed. A single text file can hold necessary flags such as paths and other info you need.
server just need to be a web server, which means IIS or similar should be installed.

I'm thinking part of your question has to do with one of the computers not having a static IP address. If the two computers, A & B, don't know each others IP address, then an alternative is to use a server. Either A can store the information on the server to be picked up by B, or A can register his current IP address on the server for B to lookup and then connect to A, assuming firewalls don't interfere. Another scheme is that A and B can simultaneously connect to the server, and the server can relay data between the two.
Of course, all of this communication would be done using techniques such as those suggested by mafutrct and Colin.

This question sounds to me like the difficulties of NAT Traversal and trying to establish peer-to-peer connectivity over the Internet. I stumbled on this question researching the best way to set up Internet connectivity to a device that is sitting behind a firewall. I am assuming that the device behind the firewall has to initiate the connection, that a Internet server application (on a Public IP address) to at least manage the initiation of the connection is required, and the Internet server application may also have to act as a relay if the difficulties of NAT is not able to be traversed.
http://en.wikipedia.org/wiki/NAT_traversal
http://en.wikipedia.org/wiki/Peer-to-peer
A good example of this is LogMeIn. Where the application is installed on the computer that needs to be accessible over the Internet, that application communicates with the LogMeIn Internet servers, you can establish a connection to that computer either by initiating through the LogMeIn servers and connecting peer-to-peer over port 80/443 using NAT Traversal (this would be establishing a peer-to-peer connection) or by initiating through the LogMeIn server and the LogMeIn server acting as a relay if peer-to-peer is unable to be established.
Please note; you must have an application (something running) on both ends of the connection on the Internet, there is no other way. Just like FTP that was mentioned earlier (FTP Client, FTP Server)... but in this case it is peer-to-peer (basically you are writing your own Client and Server... or if you have to use a relay... Client--Relay--Server
http://en.wikipedia.org/wiki/LogMeIn
"The service connects the remote desktop and the local computer using SSL over TCP or UDP and utilizing NAT traversal techniques to achieve peer-to-peer connectivity when available."
The part I don't know is can the relay go in both directions; and I am figuring the device behind the firewall must have to constantly manage the open connection to the Internet Server that is the relay. Those are the questions I am wrestling with now.
Also, you may want to check out this post.
C# byte streams through TCP

From that time, till this date I have encounter so many problems with these four kinds of NAT that basically without an external server you cant do anything.
This is because, one computer can never find where the other one is located as suggested in the pic.
Here PC 'A' will never come to know about the port that corresponds to PC 'B' on R2B without an external server.Even more, if somehow u come to know about the ports you are still in a situation that the router wont allow you to access it if the request hasn't been made for your IP.