I need to make TCP based decentralised chat app for local network. By decentralised I mean there is no central server. Each entity on a network should have server/client architecture. When app starts it should check which user is online ( already running the app ). My question is how can i check that? Can i do it by trying to connect via connect() function from socket library? I'm new to programming, especially socket programing, so if it's a dumb question sorry in advance.
You should definitely study how other decentralized applications do this. There are lots of techniques.
Each instance of the application should, as part of its server functionality, track the addresses of other instances of the application. Each instance should, as part of its client functionality, keep track of a few instances it can connect to. Prefer instances that have been around for a long time.
The software should include a list of servers that have been running for a long time and are expected to typically be available. You may wish to include a fallback method such DNS, maintained by anyone willing to keep a list of well-known servers offering access through a well-known port. The fallback method can also be IRC or HTTP.
If you want to stay decentralized, you might want to try multicasting or broadcasting a request packet to all hosts on the network to discover other instances of your chat application.
Something similar has been implemented in Pidgin, named Bonjour. It works quite nicely and provides chatting capabilities on a local network. More specifically, it is defined as a Serverless Messaging part of XMPP.
If you are looking for code examples, have a look at one of my projects where I use multicast to discover hosts on the local network that provide a specific service: Headers and implementation.
Related
I'm looking for the name of a protocol and example code that permits handing off IP/port connections to establish unmediated P2P after introduction through a server.
Simple example:
You and I both start chat programs that connect to chatintroduce.com (fictional server). I send you a "Hi! Wanna chat?" message. It doesn't get sent. Instead my chat program tells chatintroduce to send your chat program a request for connection. You respond to a prompt and your chat program tells chatintroduce to broker the connection. Chatintroduce establishes an initial two-way connection between us. Now, this final step is important, chatintroduce releases control and our two chat programs now talk directly to each other without any traffic through chatintroduce.
In other words, I construct packets which have your IP address and you receive them without interference from firewalls, NATs or any other technologies. In other words, true peer-to-peer connection independent of intermediate server.
I need to know what search terms to use to find appropriate technology. An RFC name would suffice. I've been searching for days without success.
I think what you are looking for is TCP/UDP hole punching which typically coordinates the P2P connection using a STUN server to determine the "capabilities" of the firewalls (e.g. is it a full cone nat? symmetric?).
https://en.wikipedia.org/wiki/Hole_punching_(networking)
We employed this at a company I worked for to create a kind of BitTorrent that could circumvent firewalls for streaming video between two peers.
Note that sometimes it is NOT possible to establish a connection without the intermediary.
What you are looking for is ICE protocol. RFC 5245. This protocol is used for connecting two peers through NAT traversal. There are some open source libraries and also some proprietary libraries for this. You can search google with ICE implementation.
You will also need to read about some additional protocols. These are used with ICE protocol. They are STUN and TURN.
For some cases you can't make P2P call 100% time. You will have to use a relay server. Like if the NAT combination of two peers are Symmetric vs Symmetric/PRC. That relay server is called TURN server.
Some technique like Port forwarding and TCP/UDP hole punching will help you to increase P2P rates.
See this answer for more information about which combination of NAT will require a relay server and which don't.
Thank you. I will be looking further into ICE, STUN, TURN, and hole-punching.
I also found n2n which looks like almost exactly what I wanted.
https://github.com/meyerd/n2n
http://xmodulo.com/configure-peer-to-peer-vpn-linux.html
With n2n, one makes a VPN with a super node that all other edge nodes know.
But once the introductions are made, the super node can be absent.
This was exactly what I wanted. I hope it works across platforms (linux, MacOS, Windows).
Again, I am still researching before implementation, so your advice was very important to me.
Thank you.
Use PJNATH. Its open source.
http://www.pjsip.org/pjnath/docs/html/
There is not much open source on NAT Traversal. As far as I know PJNATH is good.
For server you can use Google's Open source STUN and TURN server.
The server consists of several services with which a user interacts: profiles, game logics, physics.
I heard that it's a bad practice to have multiple client connections to the same server.
I'm not sure whether I will use UDP or TCP.
The services are realtime, they should reply as fast as possible so I don't want to include any additional rerouting if there are no really important reasons. So are there any reasons to rerote traffic through one external endpoint service to specific internal services in my case?
This seems to be multiple questions in one package. I will try to answer the ones I can identify as separate...
UDP vs TCP: You're saying "real-time", this usually means UDP is the right choice. However, that means having to deal with lost packets and possible re-ordering of packets. But, using UDP leaves a couple of possible delay-decreasing tricks open.
Multiple connections from a single client to a single server: This consumes resources (end-points, as it were) on both the client (probably ignorable) and on the server (possibly a problem, possibly ignorable). The advantage of using separate connections for separate concerns (profiles, physics, ...) is that when you need to separate these onto separate servers (or server farms), you don't need to update the clients, they just need to connect to other end-points, using code that's already tested.
"Re-router" (or "load balancer") needed: Probably not going to be an issue initially. However, it will probably become an issue later. Depending on your overall design and server OS, using UDP may actually become an asset here. UDP packet arrives at the load balancer, dispatched to the right backend and that could then in theory send back a reply with the source IP of the load balancer.
An alternative would be to have a "session broker". The client makes an initial connection to a well-known endpoint, says "I am a client, tell me where my profile, physics, what-have0-you servers are", the broker considers the current load, possibly the location of the client and other things that may make sense and the client then connects to the relevant backends on its own. The downside of this is that it's harder (not impossible, but harder) to silently migrate an ongoing session to a new backend, when there's a load-balancer in the way, this can be done essentially-transparently.
At the bottom, it's all about socket communications. If there is some way to get the ip of the both users, why can't the connection be directly setup between the users instead of having to go thru a server in the middle?
My 2 cents:
No one out there forces us to have a server based real-time communication model. Infact XMPP have an extension called "Serverless Messaging" which defines how to communicate over local or wide-area networks using the principles of zero-configuration networking for endpoint discovery and the syntax of XML streams and XMPP messaging for real-time communication. This method uses DNS-based Service Discovery and Multicast DNS to discover entities that support the protocol, including their IP addresses and preferred ports.
P2P chat applications have been for over a decade now. Having a server in the middle is purely a decision dependent upon your application needs. If your application can live with chats getting lost while the user was transitioning between online/offline status, then you can very well have a direct P2P model going. Similarly, there are a loads and loads of advantages (contact list management, avatars, entity discovery, presence authorization, offline messages, ....) when it comes to choosing a server based messaging model. If you try to have all this right inside your P2P based clients, they might die or under-perform because of all the work they will need to perform by themselves.
"WebSockets" were not designed for P2P/Serverless communication, rather they were designed to provide a standardized PUSH semantic over stateless HTTP protocol. In short, "WebSockets" is a standardized way replacing hacky comet, long-polling, chunked-encoding, jsonp, iframe-based and various other technique developers have been using to simulate server push over HTTP.
Named WebSockets (if someday it is fully and widely supported) could be the solution.
http://namedwebsockets.github.io/spec/
Named WebSockets are useful in a variety of collaborative local device
and local network scenarios: Discover matching peer services on the
local device and/or the local network.
Direct communication between users is possible in Peer To Peer (P2P) networks. In P2P each participant can act as client as well as server. But for P2P networks you need to write a separate program to make the communication possible.
Web Sockets let you leverage existing common browsers as clients. All depends on what is the purpose of your application and how you want to deploy it.
If there is some way to get the ip of the both users
You nailed the answer right in your question.
Most machines I use have IP address of 192.168.0.10 (or similar from 192.168. private network) and are deep, deep behind several layers of NAT. With the end of free IPv4 address pool and IPv6 nowhere near sight, this is the reality most users live. Having a stable intermediary of known, routable address helps a ton working around this issue.
WebSockets don't allow the socket to listen for connections, only to connect as a client to a server (not reverse). Technically they could make it allow this, but as far as I understand the spec doesn't currently (nor is it expected to) allow listen functionality for WebSockets.
The new WebRTC (http://www.webrtc.org/) spec looks like it might support peer-to-peer connections. I have not played with WebRTC at all so I'm not in a position to comment on it. I think it would be a bit more involved than WebSocket stuff. Maybe someone who knows WebRTC better can chime in. (Also apart from the latest version of Chrome I'm not sure if any of the other browsers really support WebRTC yet).
This is my first experience with P2P and i need some help regarding the design.
I am developing a simple messenger application. I have a directory server on which every user authenticates and announces an open port on which every user is reachable. The directory server maintains the users and the ports and I can query the directory server for any specific user. This part is done. The second part is the chat which i think should be P2P. I can start a chat as well as I can be end point of a chat (client as well as server)
What is confusing me is how do I deal with P2P? Do I create two different sockets? One on which I am listening for TCP requests for incoming connections and another one from which I would send TCP requests to start chat.
In this case do I need 3 sockets, one to talk with server and two for P2P?
If you want to go P2P, you'd better use a framework, such as JXTA for example if you are coding in Java. Creating sockets may not be enough by itself, because there are more complicated issues you need to deal with such as NAT traversal if you are operating beyond your LAN.
It seems like you have a central peer (some of server). If it has a public IP address, then you could implement a TURN-like architecture (peers communicate via this central peer). If you want direct connection between peers, you are looking a STUN solutions, but you still need a central peer to facilitate the communication.
TCP Stun is not easy. UDP is not very complicated, you just need to punch a hole in your NAT. Now, keep in mind that NAT traversal is not always possible (it depends of the NAT itself). In this case, the backup solution in a STUN one.
Is it possible to multiplex sa ocket connection?
I need to establish multiple connections to yahoo messenger and i am looking for a way to do this efficiently without having to hold a socket open for each client connection.
so far i have to use one socket for each client and this does not scale well above 50,000 connections.
oh, my solution is for a TELCO, so i need to at least hit 250,000 to 500,000 connections
i'm planing to bind multiple IP addresses to a single NIC to beat the 65k port restriction per IP address.
Please i would any help, insight i can get.
**most of my other questions on this site have gone un-answered :) **
Thanks
This is an interesting question about scaling in a serious situation.
You are essentially asking, "How do I establish N connections to an internet service, where N is >= 250,000".
The only way to do this effectively and efficiently is to cluster. You cannot do this on a single host, so you will need to be able to fragment and partition your client base into a number of different servers, so that each is only handling a subset.
The idea would be for a single server to hold open as few connections as possible (spreading out the connectivity evenly) while holding enough connections to make whatever service you're hosting viable by keeping inter-server communication to a minimum level. This will mean that any two connections that are related (such as two accounts that talk to each other a lot) will have to be on the same host.
You will need servers and network infrastructure that can handle this. You will need a subnet of ip addresses, each server will have to have stateless communication with the internet (i.e. your router will not be doing any NAT in order to not have to track 250,000+ connections).
You will have to talk to AOL. There is no way that AOL will be able to handle this level of connectivity without considering cutting your connection off. Any service of this scale would have to be negotiated with AOL so both you and they would be able to handle the connectivity.
There are i/o multiplexing technologies that you should investigate. Kqueue and epoll come to mind.
In order to write this massively concurrent and teleco grade solution, I would recommend investigating erlang. Erlang is designed for situations such as these (multi-server, massively-multi-client, massively-multithreaded telecommunications grade software). It is currently used for running Ericsson telephone exchanges.
While you can listen on a socket for multiple incoming connection requests, when the connection is established, it connects a unique port on the server to a unique port on the client. In order to multiplex a connection, you need to control both ends of the pipe and have a protocol that allows you to switch contexts from one virtual connection to another or use a stateless protocol that doesn't care about the client's identity. In the former case you'd need to implement it in the application layer so that you could reuse existing connections. In the latter case you could get by using a proxy that keeps track of which server response goes to which client. Since you're connecting to Yahoo Messenger, I don't think you'll be able to do this since it requires an authenticated connection and it assumes that each connection corresponds to a single user.
You can only multiplex multiple connections over a single socket if the other end supports such an operation.
In other words it's a function protocol - sockets don't have any native support for it.
I doubt yahoo messenger protocol has any support for it.
An alternative (to multiple IPs on a single NIC) is to design your own multiplexing protocol and have satellite servers that convert from the multiplex protocol to the yahoo protocol.
I'll trow in another approach you could consider (depending on how desperate you are).
Note that operating system TCP/IP implementations need to be general purpose, but you are only interested in a very specific use-case. So it might make sense to implement a cut-down version of TCP/IP (which only handles your use-case, but does that very well) in your application code.
For example, if you are using Linux, you could route a couple of IP addresses to a tun interface and have your application handle the IP packets for that tun interface. That way you can implement TCP/IP (optimised for your use-case) entirely in your application and avoid any operating system restriction on the number of open connections.
Of course, it's quite a bit of work doing the TCP/IP yourself, but it really depends on how desperate you are - i.e. how much hardware can you afford to throw at the problem.
500,000 arbitrary yahoo messenger connections - is your telco doing this on behalf of Yahoo? It seems like whatever solution has been in place for many years now should be scalable with the help of Moore's Law - and as far as I know all the IM clients have been pretty effective for a long time, and there's no pressing increase in demand that I can think of.
Why isn't this a reasonable problem to address with hardware plus traditional solutions?