RNTMNS Random name that make no sense

Configuring Eclipse IDE and lightty for FastCGI C development

You might want to skip my thoughts and go to configuration steps...

During a discussion with a programmer in my workplace I came to a conclusion that I don't know anything about FastCGI. Prehistory was that we were discussing ways how to make Internet application scale to N servers. Let's say we know(...hope...) that our application will be interesting to general public, so we expect high load on server. On the other hand, our application might not be so popular, so we will waste our money if we invest to many servers at time=0. So what we need is a solution, how to scale out easily and efficiently. One way to do this is FastCGI. We can start with Web server, Apps server and DB server in one physical machine (or even in one virtual machine) and later on scale out depending on our workload. To try this out I decided to learn a little bit about FastCGI. For underlying programming language I chose C because I know how to program it and it should be as fast as possible without dipping into assembler world.

Configuration - Eclipse IDE.

• I assume that you have installed and configured C/C++ perspective in Eclipse.
• Get the latest FastCGI release. At the time of writing it was "2.4.1-SNAP-0910052249" and was available for download from here
• Unpack FastCGI and do the
  1. ./configure
  2. make
  3. make install
• Check that installation went well. /usr/local/include should have FastCGI headers.
• Add libfcgi to linker libraries list in Eclipse project properties.
  

  fcgi library in libraries list

• In case your web server does not have libfcgi installed, be sure to click "No shared libraries" checkbox in GCC C Linker General options screen
• Compile the program from the example at the end of this post.
• If everything goes well - copy this program to your development server

Configuration - lighttpd server

• Open lighttpd.conf file in any text editor as long as it is vi(m)
• Make sure that "mod_fastcgi" is in server.modules list.
• Configure the fastcgi.server section. Example configuration:
  

  fastcgi.server =
  (
  ".php" =>
  ( "localhost" =>
  (
  "socket" => "/var/run/lighttpd/php-fastcgi.socket",
  "bin-path" => "/usr/bin/php-cgi"
  )
  )
  ,
  "/texas-cgi" =>
  ( "texas-cgi.handler" =>
  (
  "socket" => "/tmp/texas-cgi.socket",
  "bin-path" => "/srv/www/htdocs/cgi-bin/texas-cgi",
  "check-local" => "disable",
  "min-proc" => 1,
  "max-proc" => 1,
  "idle-timeout" => 100
  )
  )
  )

  Edit this part to suit your needs. Having socket in /tmp folder might not be the best idea. Be sure to change it...

• Restart the lightty /etc/init.d/lighttpd restart
• Check the lightty error log if something goes wrong
• This is how processes tree looks like in CentOS server with lightty and php
• Access the cgi page. My anonymised output can be seen in screenshot

htop showing php and cgi

Output from my firs FastCGI program

That's it folks. Get into programming C and running your programs as web pages! And stay tuned for updates!

#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <syslog.h>
#include <alloca.h>
#include <sys/types.h>
#include <unistd.h>
#include <fcgiapp.h>
#define LISTENSOCK_FILENO 0
#define LISTENSOCK_FLAGS 0
int times_run;
int main(int argc, char** argv) {
  openlog ("testfastcgi", LOG_CONS|LOG_NDELAY, LOG_USER);
  int err = FCGX_Init (); /* call before Accept in multithreaded apps */
  if (err) { syslog (LOG_INFO, "FCGX_Init failed: %d", err); return 1; }
  times_run=0;
  FCGX_Request cgi;
  err = FCGX_InitRequest(&cgi, LISTENSOCK_FILENO, LISTENSOCK_FLAGS);
  if (err) { syslog (LOG_INFO, "FCGX_InitRequest failed: %d", err); return 2; }
  while (1) {
    err = FCGX_Accept_r(&cgi);
    if (err) { syslog (LOG_INFO, "FCGX_Accept_r stopped: %d", err); break; }
    char** envp;
    int size = 200;
    for (envp = cgi.envp; *envp; ++envp) size += strlen(*envp) + 11;
    char*  result = (char*) alloca(size);
    strcpy (result, "Status: 200 OK\r\nContent-Type: text/html\r\n\r\n");
    strcat (result, "<html><head><title>testcgi</title></head><body><ul>\r\n");
    for (envp = cgi.envp; *envp; ++envp) {
      strcat(result, "<li>");
      strcat(result, *envp);
      strcat(result, "</li>\r\n");
    }
    strcat(result, "</ul>\r\n");
    int pid;
    char str[10];
    pid = getpid();
    strcat(result,"<p><b>My PID is: ");
    sprintf(str, "%d", pid);
    strcat(result, str);
    strcat(result, " And I've been run ");
    sprintf(str, "%d", times_run);
    times_run++;
    strcat(result, str);
    strcat(result, "times</b></p>\r\n");
    strcat(result, "</body></html>\r\n");
    FCGX_PutStr(result, strlen(result), cgi.out);
    FCGX_Finish_r(&cgi);
  }
  FCGX_ShutdownPending();
  return 0;
}

(c) Whoever owns it. I made this program by combining several examples from different web sites.