Skip Menu |

Preferred bug tracker

Please visit the preferred bug tracker to report your issue.

This queue is for tickets about the Workflow CPAN distribution.

Report information
The Basics
Id: 53909
Status: resolved
Priority: 0/
Queue: Workflow
People
Owner: jonasbn [...] cpan.org
Requestors: ANDK [...] cpan.org
Cc:
AdminCc:
Bug Information
Severity: Important
Broken in: 1.33_8
Fixed in: 1.34

History Show all quoted text
  Fri Jan 22 03:48:28 2010 ANDK [...] cpan.org - Ticket created
Subject: bleadperl fafafbaf breaks JONASBN/Workflow-1.33_8.tar.gz
You may have noticed that bleadperl sends mostly FAIL reports. I did a bisect and the bisect came to the end that this commit has something to do with the fails. All perls before pass the tests all after it fail, so it seems. I haven't looked closer and probably won't because after the bisect the experts must have a look. If you can, please draw your conclusions and if you believe this should be a blocker for 5.12, please bring it to perl5-porter's attention. Here is the full message from bisect: fafafbaf705adfdf05ff97b6ab5149ba4ee3c039 is first bad commit commit fafafbaf705adfdf05ff97b6ab5149ba4ee3c039 Author: Rick Delaney <rick@consumercontact.com> Date: Sun Jan 6 09:14:39 2008 -0500 Big slowdown in 5.10 @_ parameter passing Message-ID: <20080106191439.GF13935@bort.ca> p4raw-id: //depot/perl@32891 Let me know if you need further informations, Regards from Berlin,
  Sat Jan 30 15:40:29 2010 jonasbn [...] cpan.org - Taken
  Sat Feb 13 03:32:17 2010 ANDK [...] cpan.org - Dependency by ticket #54584 added
  Sat Feb 13 09:04:22 2010 ANDK [...] cpan.org - Correspondence added
Several people have noted that they cannot reproduce the bisect. I suppose the reason is that I did the bisect in November with 1.33_8 and 1.33 was released in the meantime. 1.33 with a lower version number was released after 1.33_8, so all installed Workflow version 1.33_1 .. 1.33_8 must be uninstalled to produce comparable results. This is a logistical nightmare and unlikely to produce reliable results. I would hope we get a new release, 1.34, so we can produce unbiased test results and find out if bleadperl and Workflow can work with each other in all configurations? Thanks,
  Sat Feb 13 09:04:25 2010 ANDK [...] cpan.org - Status changed from 'new' to 'open'
  Sat Feb 13 23:55:39 2010 ANDK [...] cpan.org - Correspondence added
OK, I could reproduce the bisect with Workflow 1.33, not only with the older 1.33_8. The bisect points again to fafafbaf. And I could observe that threaded bleadperls have no problem with Workflow 1.33 so far. And Nicholas observed that fafafbaf is included in 5.10.1 and there we have no problem either, no matter whether threaded or not. This means that fafafbaf only has an adverse effect on threaded bleadperl. There must be another fact in bleadperl that doesn't mix well with fafafbaf but only if threads are enabled.
  Sat Feb 13 23:58:32 2010 ANDK [...] cpan.org - Correspondence added
On 2010-02-13 23:55:39, ANDK wrote: Show quoted text
> This means > that fafafbaf only has an adverse effect on threaded bleadperl.
...on unthreaded bleadperl. Show quoted text
> There > must be another fact in bleadperl that doesn't mix well with fafafbaf > but only if threads are enabled.
...only if threads are disabled. Sorry.
  Sun Feb 14 05:09:17 2010 ANDK [...] cpan.org - Correspondence added
More fun with the t/workflow.t test. My observations now consistently show this behaviour: 5.10.0: always passes test, even in debugger 5.10.1: always passes tests but not when run in the debugger bleadperl before "fafafbaf" commit: always passes test, even in debugger bleadperl after "fafafbaf" commit with threaded perl: always passes tests but not when run in the debugger. bleadperl after "fafafbaf" commit with unthreaded perl: never passes test
  Thu Aug 05 10:28:47 2010 jonasbn [...] cpan.org - Correspondence added
Hi Andreas, Sorry for all the trouble my beta releases gave you - I clearly misunderstood the concept :) Anyways, I am looking into the issue for workflow now. I have received lots of test results from CPAN testers, indicating the problem you reported and it seems to be all releases of Perl above 5.10.0, well or so I thought, until I had a look at: http://matrix.cpantesters.org/?dist=Workflow+1.33 The issue is pretty much beyond me, but I will investigate, any pointers or suggestions are most welcome. Thanks for the effort, jonasbn
  Fri Aug 06 02:32:54 2010 jonasbn [...] cpan.org - Correspondence added
Implemented work-around for RT #53909 https://rt.cpan.org/Ticket/Display.html?id=53909 The issue demonstrated here, which can be observed in perl versions newer than 5.10.0 seems to be related to a issue mentioned here: http://rt.perl.org/rt3/Public/Bug/Display.html?id=70171 RT #53909 is based on blead perl, in which also a fix has now been implemented, but we still have issues with a lot of perl releases currently out there, see: http://matrix.cpantesters.org/?dist=Workflow+1.33 So this work-around seems to fix the issue, since I can no longer replicate the error. The problem seem to be the clearing of a package scoped variable, the array @observations in t/SomeObserver.pm
  Fri Aug 06 02:32:55 2010 jonasbn [...] cpan.org - Status changed from 'open' to 'resolved'
  Fri Aug 06 02:47:10 2010 jonasbn [...] cpan.org - Reference by ticket #54739 added
  Sat Oct 27 09:00:25 2012 jonasbn [...] cpan.org - Fixed in 1.34 added