I understand the purpose of FastCGI in terms of performance over other gateway interfaces. But if libraries that implement FastCGI already have to go through the painstaking measures of implementing a secure and efficient TCP service, why not just write applications as a web server? Is it less efficient for a front-end web server to implement a reverse proxy than it is for FCGI? Or is the specification for FCGI that much simpler than that of HTTP?
I had a web server program that I replaced with an FCGI program. Part of the reason I did it was because of the requirements of the program. It needed to run under the existing web server since doing so would require no additional configuration.
Another requirement was that it would be able to track some state values that can change frequently. The existing program wasn't very efficient, tracking these state values externally and doing a fresh load of them for every request. One advantage of an FCGI program is that it is persistent, so it could poll those state values and they would be available and ready for every request.
I didn't have to implement any of the FCGI spec, I just used an existing FCGI library and it handled all the communication with the web server for me. Compared to other applications I've written that have an embedded web server, implementing one with the FCGI library was relatively simple.
Another advantage of FCGI is that an FCGI application doesn't need to actually reside on a disk accessible to the web server. You can configure the server to establish a TCP connection to an instance of your FCGI application on another server. If you are in a situation where you can't access the web server directly, this can be very useful. Accomplishing the same thing with other methods can be cumbersome, but an FCGI application doesn't even need to be modified. Just configure the server, launch your app under the FCGI host, and you are good to go.
Yet another advantage of FCGI is that you can configure the web server to distribute requests between a preset number of instances of your application. If your web application works best by handling every request synchronously, FCGI is ideal because you can funnel every request into a single instance, and that instance can handle each request in a single loop. If you want at most 2, 3, or n instances all you need to do is change the value on the server and you can accomplish that.
So, FCGI isn't the best for every situation. It's the times you are faced with unusual requirements that FCGI is most attractive.
Related
Please read before tagging as duplicate.
I'm creating a set of applications which rely on smart cards for authentication. Up to now, each application has controlled the smart card reader individually. In a few weeks, some of my customers will be using more than one application at the same time. So, I thought maybe it would be more practical to create a service application which controls the authentication process. I'd like my desktop applications to tell the service application they are interested in the authentication process, and the service application would then provide them with information about current user. This part is easy, using named pipes. The hard part is, how can the service tell the desktop applications that an event has occurred (UserLogIn, UserLogOut, PermissionsChanged, ... to name a few). So far I have two methods in mind. CallBack functions, and Messages. Does anyone have a better idea? I'm sure someone has.
You want do to IPC (Inter Process Communication) with Delphi.
There are many links that can help you, Cromis IPC is just one to give you an idea what you are after.
A similar SO question to yours is here.
If you want to go pure Windows API, then take a look at how OutputDebugString communications is implemented.
Several tools can listen to the mechanism and many apps can send information to it.
Search for DBWIN_DATA_READY and DbWin32 for more information on how the protocol for OutputDebugString works.
This and this are good reading.
When it gets into IPC, some tips:
Do not be tied on one protocol: for instance, if you implements named pipe communication, you would later perhaps need to run it over a network, or even over HTTP;
Do not reinvent the wheel, nor use proprietary messages, but standard formats (like XML / JSON / BSON);
Callbacks events are somewhat difficult to implement, since the common pattern could be to implement a server for each Desktop client, to receive notifications from the server.
My recommendation is not to use callbacks, but polling on a stateless architecture, on the Desktop applications. You open a communication channel with the server, then every second / half second (use a TTimer in your UI), you make a small request asking for what did change (you can put a revision number or a time stamp of your last retrieval). Therefore, you synchronize your desktop data with pending events. Asking for updates on an existing connection is very fast, and will just send one IP packet over the network back and forth, if nothing changed. It is a very small task, and won't slow down nor the client nor the server (if you use some in-memory cache).
On practice, with real application, such a stateless architecture is very responsive, from the end-user point of view, and is much more easy to deploy. You do not need to create a server on each desktop application, so you don't have to open firewall ports or such. Since HTTP is stateless, it is even Internet friendly.
If you want to develop services, you can use DataSnap, something like RemObjects or you can try our Open Source mORmot framework which is able to create interface-based services with light JSON messages over REST, either in-process, using GDI messages, named pipes or TCP/HTTP - for free, with unbeatable performance, build-in security, and from Delphi 6 up to XE2. For your event-based task, just using the Client-Server ORM available in mORMot could be enough: create a table/class storing the events (you can even define a round-robin in-memory storage - no need to use SQLite3 engine nor a DB here), then ask for all pending events since the last refresh. And the server can safely be a background service, or a normal application - some mORMot users even have the same executable able to be either a stand-alone application, a server service, an application server, or a UI client, just by changing the configuration.
Edit / announcement:
On the mORMot roadmap, we added a new upcoming feature, to easily implement one-way callbacks from the server.
That is, add transparent "push" mode to our Service Oriented Architecture framework.
Aim is to implement notification events triggered from the server side, very easily from Delphi code, via some interface definitions, even over a single HTTP connection - for instance, WCF does not allow this: it will need a dual binding, so will need to open a firewall port and such.
It will used for easy Event Collaboration, via a publish / subscribe pattern, and allow Event Sourcing. I will try to make it implement the two modes: polling and lock-and-wait. A direct answer to your question.
You can use a simple TCP socket connection (bidirectional) to allow asynchronous server to client messages on the same socket.
An example is the Indy TIdTelnetClient class, it uses a thread for incoming messages from the server.
You can build a similar text-based protocol and only need a Indy TCP server instance in the service, and one Indy Client instance in the application(s).
I'm a hobby programmer trying to build a client/server application suite, using Delphi XE.
I write stand-alone applications occasionally, for my personal use. The problem is I don't know ANYTHING when it comes to networking / multi-tier. I am willing to learn though.
I looked for ways to do this and DataSnap and IntraWeb jump out as the most likely candidates for the job. But, before I start using one or the other, I would like to know a few info I wasn't able to find on the web:
If I build an IW stand-alone app and deploy it on my site (as server) will I be able to connect to it/retrieve data with a client application? I need to do that within my Delphi written app, without using a browser and without using HTML (I know very little HTML and I'm not eager to learn). I know IntraWeb isn't meant for this, but will it work?
If I go with DataSnap and build a server app, will I be able to deploy it to my website? Or do I have to make a computer on the network the server?
In case anyone wants to know, I want to connect 5 computers(from different cities), to exchange information between them, if at all possible using ADO, as I'm very familiar with it.
Intraweb is a server based tool for building web sites, accessed via a browser. It is not "proper" multi-tier, and it sounds like it isn't what you need.
DataSnap is designed to do what you want I think. First, the ADO part would be on the server, and you would define an interface that the server offers to do tasks. Then your clients would call that interface over the internet to get data or do those tasks etc.
As for whether they will work on your web server, that depends on what your server is, and the access that is provided to it. A web server is just a PC, but typical hosting is fairly restricted to running "scripted" languages on that server, which reside in the hosting directory. A Delphi server would be a running executable (usually a Windows Service), listening on a port, and needs much more access to the computer than typical. You would need at least a virtual Windows server with administrator access to the desktop to allow it to work. In the first instance though, you could happily run the server on a DSL line or similar and make it work just fine. You can worry about hosting it on the net when you have customers paying for it.
On non-web based chat system the server distinguishes its clients by their PIDs, right? And what should be used to distinguish the clients on web-based chat system?
Thnx in advance
The fact that you're using a web server shouldn't change much about your model. You're still building chat. You also don't want to make your chats tied too deeply to the process that is managing their HTTP connection. HTTP connections are ephemeral, even if everything is going well and you're using long polling there's no guarantee that the connection will be re-used with Keep-Alive for the next long poll. The user might also want to open up the same chat in multiple browser windows, multiple computers, whatever.
I haven't looked closely at any of these but you're not the first person that has built web chat with Erlang:
http://chrismoos.com/2009/09/28/building-an-erlang-chat-server-with-comet-part-1/
http://www.erlang-factory.com/upload/presentations/31/EugeneLetuchy-ErlangatFacebook.pdf
http://yoan.dosimple.ch/blog/2008/05/15/
https://github.com/yrashk/socket.io-erlang (more of a general tool for this sort of thing, not chat specifically)
https://github.com/rvirding/chat_demo (as seen above)
I think the confusion comes from the notion that a Erlang server process must stay alive for every individual client. It can, but Mochiweb doesn't do that by default if I'm not mistaken. It just spawns a new process for every request. If you would like to have a long lived bidirectional client <-> server process connection you can do that for example by;
sending a client identifier with every request and map that to a long-lived process on the server. The process will maintain servers state and you can call methods on it. It's still pull and not push though.
use the web socket implementations. Not sure if Mochiweb has one, but other Erlang HTTP servers like Misultin and Yaws provide one. For a web based chat system I believe web sockets would be a great fit.
For a very trivial example of a web-based chat system using websockets and Misultin you can check out this chat demo. It was written to demonstrate an idea and is not very elegant, but it does work.
I have a Client/Server application written Delphi. Essentially all the application is doing is transferring xml data streams between a server application and connected clients. I am currently using the Indy TIdTCPServer component. But the server side application keeps crashing on some of my installments. And it has been extremely difficult to debug. So I am wondering if there is some "architecture" I should be utilizing which does all the tcp/ip connection management and database connection pooling, allowing me to concentrate on the business logic.
Here are more details:
clients must maintain a "persistent" connection. There are times when the server must notify and send data to all connected clients.
clients are connecting from laptop computers using wireless aircards. So network "drops" are pretty common.
Backend database is SqlServer.
There can be upward of 100 computers simultaneously connected at a time.
When the server gets a new connection (TCPServer.OnConnect) I instantiate my own object containing it own SqlServer database connection. When tcp connections are dropped I in turn free these objects (and associated database connection).
Client application have a TTimer built into them. They routinely send heartbeats to the server. And if they "drop"/"lose" their connection they automatically establish a new connection once the network is back.
Anyone have any suggestions on the best approach/architecture here?
I presume the Indy component would work, but at the same time feel I am "reinventing the wheel" with respect to managing the connections.
Three component sets I am aware of that will take care of the nitty gritty technical aspects of client server applications for you:
kbmMW: http://components4developers.com/
Asta: http://www.astatech.com/index.asp
RemObjects: http://www.remobjects.com/
You may have to rework your applications to take advantage of the way these component sets work, but assuming you have properly separated layers that shouldn't be too much of a hassle and will buy you the advantage of well tested and widely used code for your client server work.
If you want some light TCP/IP components, take a look at our SynCrtSock unit.
You'll find low-level classes to create IP Client and Servers.
We implemented both TCP/IP and UDP/IP in one of our applications.
There is also a THttpServer class, which implement a HTTP/1.1 server. Therefore it follows the HTTP/1.1 connection management. There is also an optional compression, and using HTTP/1.1 on a port other than 80 is not a bad idea. And what is good with HTTP/1.1 is that it can pass through firewalls, and can be easily be VPNed or hosted on another HTTP server (like IIS or Apache) with a proxy. There is even a FastCGI class, if you need such a server under a linux-based solution.
Of course, a THttpClientSocket class does the same on the client class.
We use these classes to add HTTP/1.1 connection to our Open Source SQLite3 RESTful framework - http://synopse.info/forum/viewforum.php?id=2
See http://synopse.info/fossil/artifact?name=722e896e3d7aad1fe217b0e2e7903483e66d66d1 for the SynCrtSock unit. Open source, work from Delphi 7 to Delphi 2010.
Misha Charrett's CSI Application Framework covers pretty much exactly what you're asking for.
It's an open source Delphi framework that at its heart is a distributed message passing and threading framework that allows XML message passing from both client to server and server to client.
It can handle disconnections/reconnections, high client numbers and there's an optional virtual database library that will handle SQL server (or you could just use same SQL Server access you're using now).
It's not particularly well known yet but I can tell you that it's been actively developed over the last few years and that the author Misha is very keen to assist anyone who's interested in using it in their application.
Well, it would probably require a complete rewrite of much of your C/S code, but instead of using the Indy components, you could try to use a COM+ solution instead. Basically, you would create a COM+ component that will be installed on the server and your client applications will connect to this client and call the functions of this component directly. It will have transaction management which will be handled by Windows itself and the same is true about handling transactions. It's also technically possible to create events, which would allow the server to do callbacks to the client, although that would make things a bit more complicated.
I don't think this solution would work out for you, though, unless you have a lot of experience with COM development in Windows and/or you're brave enough to try something different.
In the past, I had a similar problem where hundreds of clients had to connect to a single server, doing all kinds of database transactions. It has a steep learning curve but me and my team managed to get things working and once we understood the technique, it resulted in a very stable and reliable solution which did manage to have up to 500 users simultaneously doing updates and other actions in a one-time extreme stress-test. But again, the learning curse is steep, so it might not be the solution you're looking for.
(Still, COM+ will use a lot of functionality that's build-in into Windows, like transaction management, database pooling and whatever more.)
If you use Indy each connection will equal a thread.
Anyway, I suggest for connecting to MSSQL to use SDAC from Devart http://www.devart.com/sdac/ and for the connection layer to use HPScktSrvr based on I/O Completion Port from http://www.torry.net/authorsmore.php?id=7131 (I don't know though what changes it will need for TThread changes in newer VCL).
You build your client class arround THPServerClient, you set your new class as the server ClientClass and the framework will create automatically new clients for you.
You may also want to have a look at the ICS/Midware combo: http://www.overbyte.be/
I am putting together a REST API and as I'm unsure how it will scale or what the demand for it will be, I'd like to be able to rate limit uses of it as well as to be able to temporarily refuse requests when the box is over capacity or if there is some kind of slashdotted scenario.
I'd also like to be able to gracefully bring the service down temporarily (while giving clients results that indicate the main service is offline for a bit) when/if I need to scale the service by adding more capacity.
Are there any best practices for this kind of thing? Implementation is Rails with mysql.
This is all done with outer webserver, which listens to the world (i recommend nginx or lighttpd).
Regarding rate limits, nginx is able to limit, i.e. 50 req/minute per each IP, all over get 503 page, which you can customize.
Regarding expected temporary down, in rails world this is done via special maintainance.html page. There is some kind of automation that creates or symlinks that file when rails app servers go down. I'd recommend relying not on file presence, but on actual availability of app server.
But really you are able to start/stop services without losing any connections at all. I.e. you can run separate instance of app server on different UNIX socket/IP port and have balancer (nginx/lighty/haproxy) use that new instance too. Then you shut down old instance and all clients are served with only new one. No connection lost. Of course this scenario is not always possible, depends on type of change you introduced in new version.
haproxy is a balancer-only solution. It can extremely efficiently balance requests to app servers in your farm.
For quite big service you end-up with something like:
api.domain resolving to round-robin N balancers
each balancer proxies requests to M webservers for static and P app servers for dynamic content. Oh well your REST API don't have static files, does it?
For quite small service (under 2K rps) all balancing is done inside one-two webservers.
Good answers already - if you don't want to implement the limiter yourself, there are also solutions like 3scale (http://www.3scale.net) which does rate limiting, analytics etc. for APIs. It works using a plugin (see here for the ruby api plugin) which hooks into the 3scale architecture. You can also use it via varnish and have varnish act as a rate limiting proxy.
I'd recommend implementing the rate limits outside of your application since otherwise the high traffic will still have the effect of killing your app. One good solution is to implement it as part of your apache proxy, with something like mod_evasive