http://www.perlmonks.org?node_id=1049917

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

Hello. Following code produce segmentation fault: 
#!/usr/bin/env perl

use v5.16.3;
use threads (exit => 'threads_only');
use threads::shared;
use FCGI;

threads->create(\&thread);
threads->create(\&thread);
threads->create(\&thread);

while (threads->list(threads::all)) { sleep 1 }    #wait until has thr
+eads

sub thread {
    sleep 3;

    threads->detach;
    threads->exit;

    return;
}

1;
__END__

[download]
but if i require FCGI first time in thread, such as: 
#!/usr/bin/env perl

use v5.16.3;
use threads (exit => 'threads_only');
use threads::shared;

threads->create(\&thread);
threads->create(\&thread);
threads->create(\&thread);

while (threads->list(threads::all)) { sleep 1 }    #wait until has thr
+eads

sub thread {
    require FCGI; # <--- HERE IS THE DIFFERENCE

    sleep 3;

    threads->detach;
    threads->exit;

    return;
}

1;
__END__

[download]
thread can exit without any errors.

This bug prevent use FCGI.pm with threads. Does somebody has a working solution to use FCGI in main thread?

Replies are listed 'Best First'.
Re: [FCGI.pm] - not threads-safe?
by BrowserUk (Patriarch) on Aug 19, 2013 at 19:03 UTC
    This bug prevent use FCGI.pm with threads. Does somebody has a working solution to use FCGI in main thread?

    It isn't a bug. It means that FCGI is not thread-safe.

    And given what FCGI does:

    Instead of creating a new process for each request, FastCGI uses persistent processes to handle a series of requests. These processes are owned by the FastCGI server, not the web server. 1

    To service an incoming request, the web server sends environment information and the page request itself to a FastCGI process over a socket (in the case of local FastCGI processes on the web server) or TCP connection (for remote FastCGI processes in a server farm). Responses are returned from the process to the web server over the same connection, and the web server subsequently delivers that response to the end-user. The connection may be closed at the end of a response, but both the web server and the FastCGI service processes persist.2

    Each individual FastCGI process can handle many requests over its lifetime, thereby avoiding the overhead of per-request process creation and termination. Processing of multiple requests simultaneously can be achieved in several ways: by using a single connection with internal multiplexing (i.e. multiple requests over a single connection); by using multiple connections; or by a combination of these techniques. Multiple FastCGI servers can be configured, increasing stability and scalability.

    it almost certainly could never be thread-safe.

    If you describe why you think you need threads, maybe we can help.

    With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
    Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
    "Science is about questioning the status quo. Questioning authority".
    In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]

      I need FCGI.pm in threads because this is a single way to create preforking fcgi server, which can process parallel requests under window.

      FCGI.pm just listen socket, accept connection and populate the environment hash.

      Why it could never be threads safe? It's perfectly work in threads under MSWin and Linux, except segfault when thread has been destroyed.

[reply]
        It's perfectly work in threads ... except segfault when thread

        Tada!

        With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
        Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
        "Science is about questioning the status quo. Questioning authority".
        In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]
Re: [FCGI.pm] - not threads-safe?
by sundialsvc4 (Abbot) on Aug 19, 2013 at 18:54 UTC

    Ping?!   (Interested party.)

    It certainly seems that you have found a significant issue, and a workaround to it.   Will anyone more knowledgeable than I, please comment as to the performance (etc ...) implications of this ... of what this code vs. that code causes the Perl interpreter to do?   Maybe of why this is happening in the first place?

Re: [FCGI.pm] - not threads-safe?
by Anonymous Monk on Aug 20, 2013 at 21:18 UTC
[reply]
      This:
      use Thread;

      is a dead duck since 5.8.0.

      With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
      Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
      "Science is about questioning the status quo. Questioning authority".
      In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]

        use Thread; is a dead duck since 5.8.0.

        Yes, but https://metacpan.org/module/Thread Ricardo SIGNES / perl-5.18.1 / Thread says its been reworked using threads:: for old code

        The working pattern for the OP is following the same pattern., and the code uses perl_mutex

[reply]

Back to Seekers of Perl Wisdom