Bug#539055: libgtk2-perl: FTBFS: Test failure

Ryan Niebur ryanryan52 at gmail.com
Fri Aug 7 06:59:48 UTC 2009


On Fri, Aug 07, 2009 at 09:38:56AM +0300, Damyan Ivanov wrote:
> -=| Daniel Schepler, Thu, Aug 06, 2009 at 06:30:18PM -0700 |=-
> > On Tuesday 28 July 2009 23:29:32 Ryan Niebur wrote:
> > > On Tue, Jul 28, 2009 at 11:27:49AM -0700, Daniel Schepler wrote:
> > > > Package: libgtk2-perl
> > > > Version: 1:1.221-1
> > > > Severity: serious
> > > >
> > > > From my pbuilder build log:
> > >
> > > what architecture? could you please include this information in your
> > > bug reports in the future?
> > 
> > Sorry, it was on amd64.
> 
> Builds fine here (amd64). I think the problem is not 
> architecture-related. Tim already pointed it failed in the same way on 
> sparc buildd.
> 

the only arch it failed on in debian is sparc, and the only arch it
failed on in ubuntu is sparc, which made me assume it was a sparc
issue.

> Looking at the tests (though not sure I understand them entirely), 
> I think there is a possibility for the two entries that are compared 
> to a predefined array to be ordered differently, causing bogus test 
> failures. Strangely, this occures only some times, possibly depending 
> on machine load/timing.
> 
> Looking at the note about add_item() being asynchronous (line 52), 
> I wonder if some of the other methods is asynchronous too.
> 
> Anyone with Gtk2-foo reading?
> 

I'll fix it. I had a patch there that lool and I thought was okay to
disable. turns out the buildds were just playing tricks on us.

-- 
_________________________
Ryan Niebur
ryanryan52 at gmail.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: Digital signature
URL: <http://lists.alioth.debian.org/pipermail/pkg-perl-maintainers/attachments/20090806/ee87f4c1/attachment.pgp>


More information about the pkg-perl-maintainers mailing list