NetBSD Problem Report #52436

From htodd@i8u.org  Thu Jul 27 14:31:52 2017
Return-Path: <htodd@i8u.org>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "mail.NetBSD.org", Issuer "mail.NetBSD.org CA" (verified OK))
	by mollari.NetBSD.org (Postfix) with ESMTPS id 775807A1FC
	for <gnats-bugs@gnats.NetBSD.org>; Thu, 27 Jul 2017 14:31:52 +0000 (UTC)
Message-Id: <20170727143150.A3BD016C9AEC@chris.i8u.org>
Date: Thu, 27 Jul 2017 07:31:50 -0700 (PDT)
From: htodd@i8u.org
Reply-To: htodd@i8u.org
To: gnats-bugs@NetBSD.org
Subject: apache crashes on 8.0-beta-amd64, eating mutexes
X-Send-Pr-Version: 3.95

>Number:         52436
>Category:       kern
>Synopsis:       apache crashes on 8.0-beta-amd64, eating mutexes
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    kern-bug-people
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Thu Jul 27 14:35:00 +0000 2017
>Closed-Date:    Mon Nov 20 03:45:14 +0000 2017
>Last-Modified:  Mon Nov 20 03:45:14 +0000 2017
>Originator:     Hisashi T Fujinaka <htodd@twofifty.com>
>Release:        NetBSD 8.0_BETA
>Organization:
>Environment:


System: NetBSD chris.i8u.org 8.0_BETA NetBSD 8.0_BETA (CHRIS) #155: Wed Jul 26 06:11:33 PDT 2017 htodd@chris.i8u.org:/usr/obj/amd64/sys/arch/amd64/compile/CHRIS amd64
Architecture: x86_64
Machine: amd64
>Description:
This could be a pkg issue, but the apache (from pkgsrc) can't be restarted because of a lack of resources (no shared memory available).

Backtrace is limited because of lack of symbols:
(gdb) bt
#0  0x000070cb4d605663 in ?? ()
#1  0x000070cb4c603ce5 in ?? () from /usr/pkg/lib/php/20160303/zlib.so
#2  0x000070cb6124bae0 in ?? ()
#3  0x000070cb4c607af9 in _fini () from /usr/pkg/lib/php/20160303/zlib.so
#4  0x0000000000000000 in ?? ()

I can rebuild with symbols if you can tell me how.
>How-To-Repeat:
Mostly just running apache on an 8.0-beta-amd64 system.

>Fix:
Yes please


>Release-Note:

>Audit-Trail:
From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/52436: apache crashes on 8.0-beta-amd64, eating mutexes
Date: Thu, 27 Jul 2017 16:41:03 +0200

 I have this in my /etc/sysctl.conf:

 # Apache/Apr/mod_python need this
 kern.ipc.semmni=100
 kern.ipc.semmns=600
 kern.ipc.semmnu=300


 Martin

From: Hisashi T Fujinaka <htodd@twofifty.com>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: kern/52436: apache crashes on 8.0-beta-amd64, eating mutexes
Date: Thu, 27 Jul 2017 10:33:44 -0700 (PDT)

 On Thu, 27 Jul 2017, Martin Husemann wrote:

 > I have this in my /etc/sysctl.conf:
 >
 > # Apache/Apr/mod_python need this
 > kern.ipc.semmni=100
 > kern.ipc.semmns=600
 > kern.ipc.semmnu=300

 Might be the answer I was looking for. The errors on restart are like:

 [Thu Jul 27 10:05:00.418487 2017] [core:emerg] [pid 17248] (28)No space left on device: AH00023: Couldn't create the ssl-cache mutex AH00016: Configuration Failed

 -- 
 Hisashi T Fujinaka - htodd@twofifty.com
 BSEE + BSChem + BAEnglish + MSCS + $2.50 = coffee

From: Hisashi T Fujinaka <htodd@twofifty.com>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: kern/52436: apache crashes on 8.0-beta-amd64, eating mutexes
Date: Thu, 27 Jul 2017 19:12:03 -0700 (PDT)

 On Thu, 27 Jul 2017, Hisashi T Fujinaka wrote:

 > > # Apache/Apr/mod_python need this
 > > kern.ipc.semmni=100
 > > kern.ipc.semmns=600
 > > kern.ipc.semmnu=300

 Even after this, I still get core dumps. The log shows lots of:
 [Thu Jul 27 18:56:45.130286 2017] [core:notice] [pid 4922] AH00052: child pid 14018 exit signal Segmentation fault (11)

 And then:
 [Thu Jul 27 19:00:00.077687 2017] [mpm_prefork:notice] [pid 4922] AH00171: Graceful restart requested, doing restart
 [Thu Jul 27 19:00:00.101163 2017] [core:notice] [pid 4922] AH00060: seg fault or similar nasty error detected in the parent process
 [Thu Jul 27 19:05:00.539824 2017] [core:warn] [pid 568] AH00098: pid file /var/run/httpd.pid overwritten -- Unclean shutdown of previous Apache run?

From: Manuel Bouyer <bouyer@antioche.eu.org>
To: gnats-bugs@netbsd.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org,
        htodd@i8u.org
Subject: Re: kern/52436: apache crashes on 8.0-beta-amd64, eating mutexes
Date: Fri, 28 Jul 2017 10:01:11 +0200

 On Fri, Jul 28, 2017 at 02:15:01AM +0000, Hisashi T Fujinaka wrote:
 > The following reply was made to PR kern/52436; it has been noted by GNATS.
 > 
 > From: Hisashi T Fujinaka <htodd@twofifty.com>
 > To: gnats-bugs@NetBSD.org
 > Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
 > Subject: Re: kern/52436: apache crashes on 8.0-beta-amd64, eating mutexes
 > Date: Thu, 27 Jul 2017 19:12:03 -0700 (PDT)
 > 
 >  On Thu, 27 Jul 2017, Hisashi T Fujinaka wrote:
 >  
 >  > > # Apache/Apr/mod_python need this
 >  > > kern.ipc.semmni=100
 >  > > kern.ipc.semmns=600
 >  > > kern.ipc.semmnu=300
 >  
 >  Even after this, I still get core dumps. The log shows lots of:
 >  [Thu Jul 27 18:56:45.130286 2017] [core:notice] [pid 4922] AH00052: child pid 14018 exit signal Segmentation fault (11)
 >  
 >  And then:
 >  [Thu Jul 27 19:00:00.077687 2017] [mpm_prefork:notice] [pid 4922] AH00171: Graceful restart requested, doing restart
 >  [Thu Jul 27 19:00:00.101163 2017] [core:notice] [pid 4922] AH00060: seg fault or similar nasty error detected in the parent process
 >  [Thu Jul 27 19:05:00.539824 2017] [core:warn] [pid 568] AH00098: pid file /var/run/httpd.pid overwritten -- Unclean shutdown of previous Apache run?

 I've seen this too in the past. The order in which you load
 php, python and/or perl modules *does* matters ...

 -- 
 Manuel Bouyer <bouyer@antioche.eu.org>
      NetBSD: 26 ans d'experience feront toujours la difference
 --

State-Changed-From-To: open->closed
State-Changed-By: htodd@NetBSD.org
State-Changed-When: Mon, 20 Nov 2017 03:45:14 +0000
State-Changed-Why:
Workaround found to stop crashing. Mutex problem should be fixed too.


>Unformatted:

Home	PR Database Search