Skip Menu |

Preferred bug tracker

Please visit the preferred bug tracker to report your issue.

This queue is for tickets about the Yars CPAN distribution.

Report information
The Basics
Id: 86332
Status: resolved
Priority: 0/
Queue: Yars

People
Owner: PLICEASE [...] cpan.org
Requestors: SREZIC [...] cpan.org
Cc:
AdminCc:

Bug Information
Severity: (no value)
Broken in: (no value)
Fixed in: 0.84_02



Subject: Core dumps on freebsd8
While running the Yars 0.83_03 test suite, a couple of core dumps appeared on my system: $ ls -1 /tmp/yars.test.1023.*/* /tmp/yars.test.1023.0Fz1ND/perl.core /tmp/yars.test.1023.1NYpNU/perl.core /tmp/yars.test.1023.4shWdT/perl.core /tmp/yars.test.1023.4uPECA/perl.core /tmp/yars.test.1023.66aL5Q/perl.core /tmp/yars.test.1023.6d1NYp/perl.core /tmp/yars.test.1023.6dExlf/perl.core /tmp/yars.test.1023.8hP06N/state.txt /tmp/yars.test.1023.AZmw_1/state.txt /tmp/yars.test.1023.Cckktm/perl.core /tmp/yars.test.1023.ESCxAE/perl.core /tmp/yars.test.1023.EeVKnw/perl.core /tmp/yars.test.1023.FCi8_j/perl.core /tmp/yars.test.1023.IIVRyT/perl.core /tmp/yars.test.1023.JoDRPX/perl.core /tmp/yars.test.1023.JrgJFu/state.txt /tmp/yars.test.1023.MaV1Sn/perl.core /tmp/yars.test.1023.NWUqbf/perl.core /tmp/yars.test.1023.RBwXga/state.txt /tmp/yars.test.1023.TSd5aC/perl.core /tmp/yars.test.1023.UKuS2G/perl.core /tmp/yars.test.1023.VdTjhG/state.txt /tmp/yars.test.1023.WO77nR/state.txt /tmp/yars.test.1023._IJ4B5/state.txt /tmp/yars.test.1023.aYihF_/perl.core /tmp/yars.test.1023.c0szC1/perl.core /tmp/yars.test.1023.cU8z64/perl.core /tmp/yars.test.1023.d1NYpN/perl.core /tmp/yars.test.1023.dExlfq/state.txt /tmp/yars.test.1023.dTjhGc/perl.core /tmp/yars.test.1023.ekIHEa/state.txt /tmp/yars.test.1023.k7Twqe/perl.core /tmp/yars.test.1023.qjZ4sN/perl.core /tmp/yars.test.1023.rgJFu0/perl.core /tmp/yars.test.1023.rkS5TG/state.txt /tmp/yars.test.1023.s3DZiS/state.txt /tmp/yars.test.1023.sE2IGn/perl.core /tmp/yars.test.1023.uPECAd/perl.core /tmp/yars.test.1023.umnuCu/perl.core /tmp/yars.test.1023.vG6Ro8/perl.core /tmp/yars.test.1023.xZdnx6/state.txt /tmp/yars.test.1023.zcU8z6/state.txt /tmp/yars.test.1023.zp5z78/state.txt BTW: the test suite shouldn't use hardcoded /tmp here. On this system the TMPDIR environment variable is set to another directory. Best to use File::Temp::tempdir, which obeys all the best practices regarding temporary directories. Regards, Slaven
I removed the hard coded /tmp in 0.83_04 just now. Running the test suite on FreeBSD 9 was able to see a core dump, despite the test suite completing successfully, so I think one of the sub-processes is probably running out of memory. I think it is too aggressive about forking off too many threads. The test suite succeed?
On 2013-06-22 20:14:30, PLICEASE wrote: Show quoted text
> I removed the hard coded /tmp in 0.83_04 just now. > > Running the test suite on FreeBSD 9 was able to see a core dump, > despite the test suite completing successfully, so I think one of > the sub-processes is probably running out of memory. I think it is > too aggressive about forking off too many threads. > > The test suite succeed?
With 0.83_05 the test passes on my freebsd9 system, but not on the older freebsd8 system. See: http://matrix.cpantesters.org/?dist=Yars;os=freebsd;reports=1#sl=5,1 Regards, Slaven
On Tue Jun 25 13:29:32 2013, SREZIC wrote: Show quoted text
> With 0.83_05 the test passes on my freebsd9 system, but not on the > older freebsd8 system. See: > http://matrix.cpantesters.org/?dist=Yars;os=freebsd;reports=1#sl=5,1 > > Regards, > Slaven
Just installed FreeBSD 8 in a VM (i386) wasn't able to reproduce this. Are there any other differences that may be relevant? Thanks, Graham
On Tue Jun 25 13:29:32 2013, SREZIC wrote: Show quoted text
> On 2013-06-22 20:14:30, PLICEASE wrote:
> > I removed the hard coded /tmp in 0.83_04 just now. > > > > Running the test suite on FreeBSD 9 was able to see a core dump, > > despite the test suite completing successfully, so I think one of > > the sub-processes is probably running out of memory. I think it
> is
> > too aggressive about forking off too many threads. > > > > The test suite succeed?
> > With 0.83_05 the test passes on my freebsd9 system, but not on the > older freebsd8 system. See: > http://matrix.cpantesters.org/?dist=Yars;os=freebsd;reports=1#sl=5,1 > > Regards, > Slaven
Are you still seeing any core dumps with version 0.84_02? I have substantially reworked the test suite.
CC: SREZIC [...] cpan.org
Subject: Re: [rt.cpan.org #86332] Core dumps on freebsd8
Date: Sun, 07 Jul 2013 22:30:28 +0200
To: bug-Yars [...] rt.cpan.org
From: Slaven Rezic <slaven [...] rezic.de>
"Graham Ollis via RT" <bug-Yars@rt.cpan.org> writes: Show quoted text
> <URL: https://rt.cpan.org/Ticket/Display.html?id=86332 > > > On Tue Jun 25 13:29:32 2013, SREZIC wrote:
>> On 2013-06-22 20:14:30, PLICEASE wrote:
>> > I removed the hard coded /tmp in 0.83_04 just now. >> > >> > Running the test suite on FreeBSD 9 was able to see a core dump, >> > despite the test suite completing successfully, so I think one of >> > the sub-processes is probably running out of memory. I think it
>> is
>> > too aggressive about forking off too many threads. >> > >> > The test suite succeed?
>> >> With 0.83_05 the test passes on my freebsd9 system, but not on the >> older freebsd8 system. See: >> http://matrix.cpantesters.org/?dist=Yars;os=freebsd;reports=1#sl=5,1 >> >> Regards, >> Slaven
> > Are you still seeing any core dumps with version 0.84_02? I have substantially reworked the test suite. >
No more core dumps, it seems. Looks good now. Regards, Slaven -- Slaven Rezic - slaven <at> rezic <dot> de Berlin Perl Mongers - http://berlin.pm.org
closing this bug.