fork problem

sandrider has asked for the wisdom of the Perl Monks concerning the following question:

Hi Monks, Wisdom needed. I have this code that uses a fork so that the parent will display a web message stating the request is being processed. Below is the code that I'm having problems with.

} elsif ($url = $cgi->param('url')) {
    $session = get_session_id();
    $cache   = get_cache_handle();
    $cache->set($session, [0, ""]);

    #getting the parameters from the cgi object
    $type    = $cgi->param('type');
    $mudpit  = $cgi->param('mudpit');
    $url     = $cgi->param('url');
    $boldRed = $cgi->param('bold_red');
    $maxHits = $cgi->param('max_hits');

    if ($pid = fork) {
        $cgi->delete_all();
        $cgi->param('session', $session);
        print $cgi->redirect($cgi->self_url());
        
    } elsif (defined $pid) {
    close STDOUT;

    ($tmpDir, $resDir) = chk_dir($path);
        
    $buf = "";

        @urlList = split /\n/, $url;
        foreach (@urlList) {
            ($link, $fn) = which_svr($_);
            dl_file($link, $fn, $tmpDir);
            $buf .= "$fn\n";
            $cache->set($session, [0, $buf]);
        }

        #$cache->set($session, [1, $buf]);
        exit;
        
  } else {
    die "Cannot fork: $!";
  }
[download]

It seems my fork isn't working very well. The process seems to have to complete before anything is shown on the webpage. The webpage display part is below

if ($session = $cgi->param('session')) {

    $cache = get_cache_handle();
    $data  = $cache->get($session);
    
    print $cgi->header,
                $cgi->start_html(
                    -title => 'Extract peptide masses',
                    -style => { src => '/stylesheet.css' },
                    ($data->[0] ? () : (-head => ["<meta http-equiv=re
+fresh content=5>"]))
                ); #generate HTML document start

print '<img src="/icons/title_banner.gif">';

print << 'HTML_END' unless $data->[0];
    <h3>
        <p>The system is now getting the required file(s) and extracti
+ng the required information.</p>
        <p>Please do not close this browser.</p>
        <p>Links to the result(s) will be displayed when the process i
+s done.</p>
    </h3>

HTML_END

print $cgi->pre($cgi->escapeHTML($data->[1]));
print $cgi->end_html;
[download]

I have $|=1 in my codes. Can anyone tell me what to do to allow the "hold on" page to be displayed while the process the running? Thanks. Desmond

Comment on fork problem Select or Download Code

Replies are listed 'Best First'.
Re: fork problem by Limbic~Region (Chancellor) on Jun 28, 2005 at 12:59 UTC
sandrider, I am not responding regarding your original question but making a comment on code style you may want to consider. `#getting the parameters from the cgi object $type = $cgi->param('type'); $mudpit = $cgi->param('mudpit'); $url = $cgi->param('url'); $boldRed = $cgi->param('bold_red'); $maxHits = $cgi->param('max_hits');` [download] I would most likely have written that as: `my %data; $data{$_} = $cgi->param($_) for qw(type mudpit url bold_red max_hits);` [download] You may also want to check out the Vars method if you need the entire parameter list fetched. Cheers - L~R	[reply] [d/l] [select]
Re: fork problem by zentara (Cardinal) on Jun 28, 2005 at 12:25 UTC
I'm not the biggest expert in the world on this, but since no one else answered you, :-) here is the basic problem, as demonstrated in this old newsgroup posting. #!/usr/bin/perl use warnings; use strict; #Benjamin Goldberg #The Apache process does NOT wait for the CGI #process to exit, it waits #for the pipes to be closed. # (Note that if the CGI is a NPH script, then #it does waitpid() for the script to exit). #Thus, you do not really need a child process #-- you can have your CGI #process transmute itself into whatever you #would have the child process #do, by doing something like: $\| = 1; # need either this or to explicitly flush stdout, etc. print "Content-type: text/plain\n\n"; print "Going to start the fork now\n"; open( STDIN, "</dev/null"); open( STDOUT, ">>/dev/null"); open( STDERR, ">>/path/to/logfile"); fork and exit; exec "program that would be the child process"; #Apache waits for all children to close the pipes to apache. It #does not wait for them to actually exit -- think ... #how in the world could it possibly know that those #processes were really spawned from #the CGI process? Answer: It can't. It can only know #that something still has an open filedescriptor which #points to one of the pipes that #apache created to capture the CGI's stdout and stderr. #As long as one of these pipes is open, then apache waits. [download] I'm not really a human, but I play one on earth. flash japh	[reply] [d/l]
Re: fork problem by etcshadow (Priest) on Jun 28, 2005 at 13:51 UTC
Apache has a somewhat complicated relationship with children. In order to handle this well, you'll want to use Apache::SubProcess and its "cleanup_for_exec" subroutine (in the newly forked child) to better cut the ties between the parent and the forked child. Also, if you're running under mod_perl, you'll want to be careful to make the child "CORE::exit", rather than just "exit", since, under mod_perl, exit is really just a signal to finish the request and return the (not dead!) process to the pool of processes waiting to accept new socket connections. `------------ :Wq Not an editor command: Wq` [download]	[reply] [d/l]