Bug#551228: libgstreamer-perl: FTBFS: failed test

[Lucas Nussbaum]

On 16/10/09 at 21:10 +0200, gregor herrmann wrote:
> tag 551228 + unreproducible
> thanks
> 
> On Fri, 16 Oct 2009 18:31:37 +0200, Lucas Nussbaum wrote:
> 
> > During a rebuild of all packages in sid, your package failed to build on
> > amd64.
> 
> > > #   Failed test 'The thing isa GStreamer::Caps'
> > > #   at t/GstTypeFindFactory.t line 16.
> > > #     The thing isn't defined
> > > # Looks like you failed 1 test of 2.
> > > t/GstTypeFindFactory.t .. 
> > > Dubious, test returned 1 (wstat 256, 0x100)
> > > Failed 1/2 subtests 
> > > Failed 1/31 test programs. 1/512 subtests failed.
> > > make[1]: *** [test_dynamic] Error 255
> 
> Hm, builds fine for me on i386 and amd64 (both in a sid cowbuilder
> chroot).

That's strange: it fails reproducibly, and I can't find anything that
would cause the load on the build nodes to make it fail. Also, when I
run the build manually, it doesn't fail.

Looking at the build log, it seems that the ordering of the different
steps varies sometimes. Do you know if dh does parallel building behind
the scenes ?

On interesting fact is that I'm building on tmpfs, and that builds on
tmpfs have a "Parsing XS files..." line in the log that I can't find in
the other logs:
chmod 755 blib/arch/auto/GStreamer/GStreamer.so
cp GStreamer.bs blib/arch/auto/GStreamer/GStreamer.bs
chmod 644 blib/arch/auto/GStreamer/GStreamer.bs
Parsing XS files...
Generating POD...
Loaded 6 extra types from doctypes

But I also can build fine on tmpfs (starting the build manually).
-- 
| Lucas Nussbaum
| lucas at lucas-nussbaum.net   http://www.lucas-nussbaum.net/ |
| jabber: lucas at nussbaum.fr             GPG: 1024D/023B3F4F |