[sr-dev] Fwd: [SR-Users] SIGUSR1 for memory status not working as documented - only one process reports back

Alex Villací­s Lasso a_villacis at palosanto.com
Thu Jul 31 20:13:38 CEST 2014


El 31/07/14 10:47, Alex Villací­s Lasso escribió:
>
>
>
> -------- Mensaje original --------
> Return-Path: 	<sr-users-bounces at lists.sip-router.org>
> Delivered-To: 	<a_villacis at palosanto.com>
> Received: 	from palosanto.com by mail.palosanto.com (Dovecot) with LMTP id w6B7By5S2VNcGwAA3RMWGw for <a_villacis at palosanto.com>; Wed, 30 Jul 2014 15:15:08 -0500
> Received: 	from localhost (mail.palosanto.com [127.0.0.1]) by palosanto.com (Postfix) with ESMTP id B9FD413C0277 for <a_villacis at palosanto.com>; Wed, 30 Jul 2014 15:15:08 -0500 (ECT)
> X-Virus-Scanned: 	Debian amavisd-new at mail.palosanto.com
> X-Spam-Flag: 	NO
> X-Spam-Score: 	-2.237
> X-Spam-Level: 	
> X-Spam-Status: 	No, score=-2.237 tagged_above=-1000 required=6.31 tests=[AWL=-0.337, BAYES_00=-1.9] autolearn=ham
> Received: 	from palosanto.com ([127.0.0.1]) by localhost (mail.palosanto.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id SE504KHIt6Lb for <a_villacis at palosanto.com>; Wed, 30 Jul 2014 15:15:04 -0500 (ECT)
> Received: 	from www.kamailio.org (main.kamailio.org [193.22.119.66]) by palosanto.com (Postfix) with ESMTPS id A469B13C026F for <a_villacis at palosanto.com>; Wed, 30 Jul 2014 15:15:04 -0500 (ECT)
> Received: 	from localhost ([127.0.0.1] helo=main.kamailio.org ident=list) by www.kamailio.org with esmtp (Exim 4.72) (envelope-from <sr-users-bounces at lists.sip-router.org>) id 1XCaHb-000818-IE; Wed, 30 Jul 2014 22:15:31 +0200
> Received: 	from lab2.palosanto.com ([201.234.196.173] helo=palosanto.com) by www.kamailio.org with esmtp (Exim 4.72) (envelope-from <a_villacis at palosanto.com>) id 1XCaHZ-00080H-6l for sr-users at lists.sip-router.org; Wed, 30 Jul 2014 22:15:29 +0200
> Received: 	from localhost (mail.palosanto.com [127.0.0.1]) by palosanto.com (Postfix) with ESMTP id 42DF313C0277 for <sr-users at lists.sip-router.org>; Wed, 30 Jul 2014 15:14:53 -0500 (ECT)
> X-Virus-Scanned: 	Debian amavisd-new at mail.palosanto.com
> Received: 	from palosanto.com ([127.0.0.1]) by localhost (mail.palosanto.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id jNanXV3k3Sem for <sr-users at lists.sip-router.org>; Wed, 30 Jul 2014 15:14:50 -0500 (ECT)
> Received: 	from avillacis.palosanto.com (avillacis.palosanto.com [192.168.3.2]) by palosanto.com (Postfix) with ESMTPSA id 0529713C026F for <sr-users at lists.sip-router.org>; Wed, 30 Jul 2014 15:14:50 -0500 (ECT)
> Message-ID: 	<53D95256.9050206 at palosanto.com>
> Date: 	Wed, 30 Jul 2014 15:15:18 -0500
> From: 	Alex Villací­s Lasso <a_villacis at palosanto.com>
> User-Agent: 	Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.7.0
> MIME-Version: 	1.0
> To: 	Kamailio (SER) - Users Mailing List <sr-users at lists.sip-router.org>
> Subject: 	[SR-Users] SIGUSR1 for memory status not working as documented - only one process reports back
> X-BeenThere: 	sr-users at lists.sip-router.org
> X-Mailman-Version: 	2.1.13
> Precedence: 	list
> Reply-To: 	Kamailio (SER) - Users Mailing List <sr-users at lists.sip-router.org>
> List-Id: 	"Kamailio \(SER\) - Users Mailing List" <sr-users.lists.sip-router.org>
> List-Unsubscribe: 	<http://lists.sip-router.org/cgi-bin/mailman/options/sr-users>, <mailto:sr-users-request at lists.sip-router.org?subject=unsubscribe>
> List-Archive: 	<http://lists.sip-router.org/pipermail/sr-users>
> List-Post: 	<mailto:sr-users at lists.sip-router.org>
> List-Help: 	<mailto:sr-users-request at lists.sip-router.org?subject=help>
> List-Subscribe: 	<http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users>, <mailto:sr-users-request at lists.sip-router.org?subject=subscribe>
> Content-Transfer-Encoding: 	7bit
> Content-Type: 	text/plain; charset="us-ascii"; Format="flowed"
> Sender: 	sr-users-bounces at lists.sip-router.org
> Errors-To: 	sr-users-bounces at lists.sip-router.org
>
>
>
> I am trying to track down a memory leak that was triggered by a patch I wrote for my local copy of kamailio 4.1.4 . For this, I am following the documentation athttp://www.kamailio.org/dokuwiki/doku.php/troubleshooting:memory  . This page claims that once
> memlog is set in the configuration file, a kamailio process will dump a report of the allocation map when shutting down, or when receiving a SIGUSR1. I have configured my kamailio.cfg with memlog=1 and no other change, and I see the memory report on
> shutdown for all processes. However, when I send a SIGUSR1 to a kamailio process, the process does absolutely nothing, with the exception of the first kamailio process (the one reported as Type=attendant by "kamctl ps"). All of the other processes just
> ignore SIGUSR1. What is going on? It is inconvenient to force a shutdown of all the kamailio processes just to get the memory report.
>

 From examination of the source code, I see that the disabling of memory dumps on SIGUSR1 for child processes other than the main one is not a bug, but a design decision. Why?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sip-router.org/pipermail/sr-dev/attachments/20140731/f5f13382/attachment-0001.html>


More information about the sr-dev mailing list