Hi Daniel,

well, i think its not needed but wouldn't we all benefit from it ? binding the internal structure to the result instead of connection?

Thanks


On Mon, Sep 8, 2014 at 3:37 PM, <sr-dev-request@lists.sip-router.org> wrote:
Send sr-dev mailing list submissions to
        sr-dev@lists.sip-router.org

To subscribe or unsubscribe via the World Wide Web, visit
        http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev
or, via email, send a message with subject or body 'help' to
        sr-dev-request@lists.sip-router.org

You can reach the person managing the list at
        sr-dev-owner@lists.sip-router.org

When replying, please edit your Subject line so it is more specific
than "Re: Contents of sr-dev digest..."


Today's Topics:

   1. Re: accessing database from a child forked proces
      (Daniel-Constantin Mierla)


----------------------------------------------------------------------

Message: 1
Date: Mon, 08 Sep 2014 16:37:21 +0200
From: Daniel-Constantin Mierla <miconda@gmail.com>
To: Luis Azedo <luis.azedo@factorlusitano.com>
Cc: "Kamailio \(SER\) - Development Mailing List"
        <sr-dev@lists.sip-router.org>
Subject: Re: [sr-dev] accessing database from a child forked proces
Message-ID: <540DBF21.8070802@gmail.com>
Content-Type: text/plain; charset="utf-8"; Format="flowed"

Hello,

On 08/09/14 15:29, Luis Azedo wrote:
> Hi Daniel,
>
> all problems solved with presence! i was not running child_init  and
> now there is no need for the proposed patch to presence.
ok. Given this, would the db_text patch still be required, as there
should be no database connection sharing?

Cheers,
Daniel
>
> Thank you
>
>
>
> On Mon, Sep 8, 2014 at 2:13 PM, Luis Azedo
> <luis.azedo@factorlusitano.com <mailto:luis.azedo@factorlusitano.com>>
> wrote:
>
>     Hi Daniel,
>
>     this is the way we are creating the child processes (1 manager and
>     n workers)
>     the problem is with the workers.
>     the worker will call a event-route and in the script config we try
>     to call pres_refresh_watchers and that's where we get the pa_db =
>     NULL.
>
>     as i understand from your email, if we change PROC_NOCHLDINIT and
>     let the child_init execute for the forked process then it will
>     also execute child_init in other modules ? it makes sense.
>
>     going to try this.
>
>
>     static int mod_child_init(int rank)
>     {
>     int pid;
>     int i;
>
>     fire_init_event(rank);
>
>     if (rank==PROC_MAIN) {
>     pid=fork_process(PROC_NOCHLDINIT, "AMQP Manager", 1);
>     if (pid<0)
>     return -1; /* error */
>     if(pid==0){
>     kz_amqp_manager_loop(0);
>     }
>     else {
>     for(i=0; i < dbk_consumer_processes; i++) {
>     pid=fork_process(PROC_NOCHLDINIT, "AMQP Consumer", 1);
>     if (pid<0)
>     return -1; /* error */
>     if(pid==0){
>     mod_consumer_proc(i+1);
>     }
>     }
>     }
>     }
>
>     return 0;
>     }
>
>
>
>     On Mon, Sep 8, 2014 at 1:11 PM, Daniel-Constantin Mierla
>     <miconda@gmail.com <mailto:miconda@gmail.com>> wrote:
>
>         Hello,
>
>         the database connection should not be shared beween processes,
>         because it can bring unexpected results in may places.
>
>         Right now, the rule is to have one connection per process,
>         shared by all modules in that process.
>
>         To achieve that, at mod_init each module opens database
>         connection and closes it before ending the function. Then in
>         child_init() the connection is opened again. Another module
>         that will have to open in child_init() will get the same
>         connection now.
>
>         When you create a new process, you tell the type of child and
>         based on that child_init() from the other modules are executed.
>
>         What is the function do you use for creating a new process?
>         Maybe you can paste it here exactly how you do it and I can
>         see if something can be done.
>
>         Cheers,
>         Daniel
>
>
>         On 03/09/14 12:09, Luis Azedo wrote:
>>         Hi Jason,
>>
>>         thanks for the reply.
>>
>>         the last 2 statements in presence module mod_init close the
>>         connection and set pa_db to NULL. when my module main process
>>         forks the extra processes the pa_db in presence is NULL, so
>>         when it calls pres_refresh_watchers it fails because pa_db is
>>         NULL.
>>         i commented these last statements in presence mod_init and i
>>         got it to work.
>>
>>         // pa_dbf.close(pa_db);
>>         // pa_db = NULL;
>>
>>         does this have any implications on how the module works ? is
>>         it ok to merge this change ?
>>
>>         Thank you
>>
>>
>>
>>             ----------------------------------------------------------------------
>>
>>             Message: 1
>>             Date: Sun, 31 Aug 2014 09:40:49 +0200
>>             From: Jason Penton <jason.penton@gmail.com
>>             <mailto:jason.penton@gmail.com>>
>>             To: "Kamailio (SER) - Development Mailing List"
>>                     <sr-dev@lists.sip-router.org
>>             <mailto:sr-dev@lists.sip-router.org>>
>>             Subject: Re: [sr-dev] accessing database from a child
>>             forked process
>>             Message-ID:
>>
>>             <CALoGXNWvHhCAO91Tfa0w8W3eYQRvfV7Qkgte7dBnD+ciNr_Kpg@mail.gmail.com
>>             <mailto:CALoGXNWvHhCAO91Tfa0w8W3eYQRvfV7Qkgte7dBnD%2BciNr_Kpg@mail.gmail.com>>
>>             Content-Type: text/plain; charset="utf-8"
>>
>>             To confirm exactly what processes are being used, maybe
>>             check the log file
>>             and take note of process id at each log event. For
>>             example you could check
>>             the process id of the messages showing you the connection
>>             is null. Then run
>>             kamcmd ps to show the process list with a description of
>>             each Kamailio
>>             process. That will probably point you in the correct
>>             direction.
>>
>>             Cheers
>>             Jason
>>
>>
>>             On Fri, Aug 29, 2014 at 3:53 PM, Luis Azedo
>>             <luis.azedo@factorlusitano.com
>>             <mailto:luis.azedo@factorlusitano.com>>
>>             wrote:
>>
>>             > Hi,
>>             >
>>             > i have a module that creates 1 extra process where it
>>             processes stuff in a
>>             > loop.
>>             > on some condition i fire a route_event with a fakemsg
>>             and its up to the
>>             > user of the module to take action, it tries to work
>>             like dispatcher module
>>             > or htable (mod-init) events.
>>             >
>>             > the problem that i have is that, if i call some
>>             function on some module
>>             > that performs a database action, the connection is null
>>             when it calls
>>             > use_table.
>>             >
>>             > in this case i'm making this call
>>             > event_route[my_module:my_event]
>>             > {
>>             >  $var(my_uri) = <<result of some operations>>;
>>             > pres_refresh_watchers("$var(my_uri)", "dialog", 1);
>>             > }
>>             > presence module makes the call to use_table but this
>>             call fails because
>>             > the connection is null. presence module is working fine
>>             besides this.
>>             >
>>             > if i make this call inside
>>             event_route[dispatcher:dst-up] it works.
>>             >
>>             > i think that dispatcher fires the event inside a
>>             callback from a
>>             > registered timer, so, i think (may be wrong) that it
>>             comes from a different
>>             > process ?
>>             >
>>             > i wonder if i'm missing something from child_init ?
>>             need to register
>>             > something ?
>>             >
>>             > thanks for your help.
>>             >
>>             >
>>             >
>>             >
>>             >
>>             >
>>             > _______________________________________________
>>             > sr-dev mailing list
>>             > sr-dev@lists.sip-router.org
>>             <mailto:sr-dev@lists.sip-router.org>
>>             > http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev
>>             >
>>             >
>>             -------------- next part --------------
>>             An HTML attachment was scrubbed...
>>             URL:
>>             <http://lists.sip-router.org/pipermail/sr-dev/attachments/20140831/9fba51e4/attachment-0001.html>
>>
>>             ------------------------------
>>
>>             _______________________________________________
>>             sr-dev mailing list
>>             sr-dev@lists.sip-router.org
>>             <mailto:sr-dev@lists.sip-router.org>
>>             http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev
>>
>>
>>             End of sr-dev Digest, Vol 70, Issue 71
>>             **************************************
>>
>>
>>
>>
>>         _______________________________________________
>>         sr-dev mailing list
>>         sr-dev@lists.sip-router.org  <mailto:sr-dev@lists.sip-router.org>
>>         http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev
>
>         --
>         Daniel-Constantin Mierla
>         http://twitter.com/#!/miconda  <http://twitter.com/#%21/miconda>  -http://www.linkedin.com/in/miconda
>         Next Kamailio Advanced Trainings 2014 -http://www.asipto.com
>         Sep 22-25, Berlin, Germany
>
>
>

--
Daniel-Constantin Mierla
http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
Next Kamailio Advanced Trainings 2014 - http://www.asipto.com
Sep 22-25, Berlin, Germany

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sip-router.org/pipermail/sr-dev/attachments/20140908/6da7f048/attachment.html>

------------------------------

_______________________________________________
sr-dev mailing list
sr-dev@lists.sip-router.org
http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev


End of sr-dev Digest, Vol 71, Issue 24
**************************************