Skip Menu |

Preferred bug tracker

Please visit the preferred bug tracker to report your issue.

This queue is for tickets about the Test-Simple CPAN distribution.

Report information
The Basics
Id: 14536
Status: resolved
Priority: 0/
Queue: Test-Simple
People
Owner: Nobody in particular
Requestors: thm [...] duke.edu
Cc:
AdminCc:
Bug Information
Severity: Unimportant
Broken in: 0.60
Fixed in: (no value)

History Show all quoted text
  Fri Sep 09 10:58:34 2005 Guest - Ticket created
Subject: Test-Simple fails with NPTL
Test-Simple-0.60, pulled from CPAN on Sept 9. tests fail with a core dump of the test program (I could not find the core) Setting LD_ASSUME_KERNEL=2.2.5 caused the test to succeed. perl 5.8.5-12.1 running on centos-4 (rhel-4 clone). Please let me know what other information I should provide. Snip from running log: t/simple..................ok t/skip....................ok 7/15 skipped: Just testing the skip interface. t/skipall.................ok t/sort_bug................dubious Test returned status 255 (wstat -1, 0xffffffff) test program seems to have generated a core after all the subtests completed successfully t/thread_taint............ok t/threads.................dubious Test returned status 255 (wstat -1, 0xffffffff) test program seems to have generated a core after all the subtests completed successfully t/todo....................ok t/undef...................ok t/use_ok..................ok t/useing..................ok Failed Test Stat Wstat Total Fail Failed List of Failed ------------------------------------------------------------------------------- t/sort_bug.t 255 -1 2 0 0.00% ?? t/threads.t 255 -1 6 0 0.00% ?? 2 tests and 9 subtests skipped. Failed 2/55 test scripts, 96.36% okay. 0/410 subtests failed, 100.00% okay. make[1]: *** [test_dynamic] Error 255 make[1]: Leaving directory `/root/.cpan/build/Test-Simple-0.60' /usr/bin/make test -- NOT OK
  Mon Sep 19 19:01:44 2005 mschwern [...] cpan.org - Correspondence added
[guest - Fri Sep 9 10:58:34 2005]: Show quoted text
> Test-Simple-0.60, pulled from CPAN on Sept 9. > tests fail with a core dump of the test program (I could not find the > core)
Check your ulimit -c. If its 0, set it to unlimited and run the tests again. Show quoted text
> Setting LD_ASSUME_KERNEL=2.2.5 caused the test to succeed. > > perl 5.8.5-12.1 running on centos-4 (rhel-4 clone).
Since both those tests involve threads, you're using a vendor patched Perl, weird flags make it go away and its Redhat derived (Redhat routinely screws up Perl) I'm going to say you've tickled some threading bug in your particular version of Perl and there's not a whole lot I can do. Try compiling up a clean 5.8.5 from CPAN and see if it still segfaults. If it doesn't then its something Centos is doing wrong. PS What's NPTL?
  Mon Sep 19 19:01:45 2005 mschwern [...] cpan.org - Status changed from 'new' to 'stalled'
  Tue Sep 20 11:24:55 2005 The RT System itself - Status changed from 'stalled' to 'open'
  Tue Sep 20 11:24:55 2005 Guest - Correspondence added
From: thm [...] duke.edu
[MSCHWERN - Mon Sep 19 19:01:44 2005]: Show quoted text
> [guest - Fri Sep 9 10:58:34 2005]:
> > Test-Simple-0.60, pulled from CPAN on Sept 9. > > tests fail with a core dump of the test program (I could not find the > > core)
> > Check your ulimit -c. If its 0, set it to unlimited and run the tests > again. > >
> > Setting LD_ASSUME_KERNEL=2.2.5 caused the test to succeed. > > > > perl 5.8.5-12.1 running on centos-4 (rhel-4 clone).
> > Since both those tests involve threads, you're using a vendor patched > Perl, weird flags make it go away and its Redhat derived (Redhat > routinely screws up Perl) I'm going to say you've tickled some threading > bug in your particular version of Perl and there's not a whole lot I
can do. Show quoted text
> > Try compiling up a clean 5.8.5 from CPAN and see if it still segfaults. > If it doesn't then its something Centos is doing wrong.
Since I can bypass the bug without a complete perl rebuild, I'm not likely to do that for time reasons. Show quoted text
> > > PS What's NPTL?
Native Posix Threading Library LD_ASSUME_KERNEL=2.2.5 turns NPTL off, so I suspect the module isn't compatible with NPTL.
  Tue Sep 20 13:19:06 2005 schwern [...] pobox.com - Correspondence added
Date: Tue, 20 Sep 2005 10:18:46 -0700
From: Michael G Schwern <schwern [...] pobox.com>
To: Guest via RT <bug-Test-Simple [...] rt.cpan.org>
CC: undisclosed-recipients: ;
Subject: Re: [cpan #14536] Test-Simple fails with NPTL
RT-Send-Cc:
On Tue, Sep 20, 2005 at 11:24:55AM -0400, Guest via RT wrote: Show quoted text
> Native Posix Threading Library > > LD_ASSUME_KERNEL=2.2.5 turns NPTL off, so I suspect the module isn't > compatible with NPTL.
The module is doing nothing but the most trivial threading work to test that it doesn't blow up when someone uses it with threads. I doubt this problem is specific to Test::More. If Test::More's simple threading code is causing seg faults you're going to see problems with threads elsewhere. -- Michael G Schwern schwern@pobox.com http://www.pobox.com/~schwern Don't try the paranormal until you know what's normal. -- "Lords and Ladies" by Terry Prachett
  Sat Sep 24 01:23:48 2005 mschwern [...] cpan.org - Status changed from 'open' to 'stalled'
  Sat Sep 24 01:23:53 2005 mschwern [...] cpan.org - Severity Important changed to Unimportant
  Thu Sep 13 23:27:28 2007 mschwern [...] cpan.org - Correspondence added
No response from the reader. Can't diagnose, don't know what NPTL is. Likely a broken Perl with threads.
  Thu Sep 13 23:27:29 2007 The RT System itself - Status changed from 'stalled' to 'open'
  Thu Sep 13 23:27:30 2007 mschwern [...] cpan.org - Status changed from 'open' to 'resolved'