Link to home
Start Free TrialLog in
Avatar of m3p
m3p

asked on

How to make server "let-go" of Server-Side-Include Perl script after script sends back html (so script can continue with addl processing) and not hold-up loading of page

Here's my question:

The scenario is as follows:

- Client browser requests an HTML page containing an embedded  
  Server-Side-Include

- The server fetches the page and begins to parse its contents while
  (I assume), sending the parsed contents back to the client browser

- The server encounters the SSI and takes the necessary action

- The action in this case is an instruction to execute a CGI Perl script

- The Perl script is executed and sends back the obligatory
  "Content-type: text/html \n\n" header along with some HTML

*** HERE IS MY PROBLEM ***

- For all intents and purposes, the Perl script is done sending HTML
  back to the Server for inclusion into the page being sent to the
  client browser, BUT, it still has some processing to do.

  ???????? How do I tell the server to "let go" of the CGI Perl script
           (which will still keep running until it is done with
           its processing), and allow the rest of the page to be
           parsed and sent back to the client browser   ?????????

Any information or recommendations would be much appreciated.

Thanks in advance,

Jim Raposa
Webmaster
www.TextEx.net
Avatar of Paul Maker
Paul Maker
Flag of United Kingdom of Great Britain and Northern Ireland image

i dont think that is possible because the way a web server works is to exe the script and then send the resulting stdout back through the socket to the browser .

you could try flushing the stdout in the script once you have finished gerating user output . this will leave the last bit of the script to run but will not detach it from the user.
ASKER CERTIFIED SOLUTION
Avatar of mattrope
mattrope

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of m3p
m3p

ASKER

Thanks for at least trying to answer. I have tried "forking" but the server still ends up waiting for the forked process to finish (I admit maybe I'm not doing the forking correctly), anyway the circumstances in this case make it difficult for me to experiment because my site is hosted on a server that allows CGI but not Telnet, so debugging is difficult. Thanks again for trying.

Jim Raposa
Webmaster
www.TextEx.net
webmaster@TextEx.net
Did you fork twice?  I think a double fork is required to make it an independent process...
I also just found this on the perlipc manpage. It may also help you...



In some cases (starting server processes, for instance) you'll want to complete dissociate the child process from the parent. The easiest way is to use:


    use POSIX qw(setsid);
    setsid()            or die "Can't start a new session: $!";

However, you may not be on POSIX. The following process is reported to work on most Unixish systems. Non-Unix users should check their Your_OS::Process module for other solutions.

Open /dev/tty and use the TIOCNOTTY ioctl on it. See tty(4) for details.


Change directory to /

Reopen STDIN, STDOUT, and STDERR so they're not connected to the old tty.

Background yourself like this:

    fork && exit;

Ignore hangup signals in case you're running on a shell that doesn't automatically no-hup you:


    $SIG{HUP} = 'IGNORE';       # or whatever you'd like

Avatar of m3p

ASKER

Regarding the POSIX recommendation, where does the "use" code go and where does the "setsid()" code go, somewhere in the original Perl script after a "fork" perhaps, or in some script that the original attempts to "child off" ??????
Avatar of m3p

ASKER

Regarding the POSIX recommendation, where does the "use" code go and where does the "setsid()" code go, somewhere in the original Perl script after a "fork" perhaps, or in some script that the original attempts to "child off" ??????
I've never tried doing this, but I think that the POSIX method allows you to fork() once and then call the setsid() function from the forked code, thus separating that process from the parent.  The 'use' command can probably go anywhere (generally placed near the top of the program), the setsid() function should be called after you've forked, inside the child process code.  Hope that works!
Avatar of m3p

ASKER

I've stumbled upon the answer fellas. As it turns out, all that is required is that SSI'd Perl script does a "close (STDOUT)" after having sent back whatever html to the server, thereby causing the server to finally "let go" of the script allowing the page to continue loading and the script to proceed with whatever processing it still needs to do. thanks for all your help guys. if I can ever return the favor, just drop me an email.

Jim Raposa
Webmaster
www.TextEx.net
webmaster@TextEx.net