Bug#666601: libapache-singleton-perl: FTBFS: test failed

Niko Tyni ntyni at debian.org
Sun Apr 1 18:51:46 UTC 2012


tag 666601 - unreproducible
tag 666601 + patch
thanks

On Sun, Apr 01, 2012 at 03:38:04PM +0200, gregor herrmann wrote:
> tag 666601 + unreproducible
> thanks
> 
> On Sat, 31 Mar 2012 21:45:50 +0200, Lucas Nussbaum wrote:
> 
> > During a rebuild of all packages in sid, your package failed to build on
> > amd64.
> 
> > > #   Failed test 'printers are same instance'
> > > #   at t/apache/singleton/process/printer_instance.t line 13.
> > > #          got: 'Printer::PerProcess=HASH(0x2b2f4e8a6478)'
> > > #     expected: 'Printer::PerProcess=HASH(0x2b2f51a727c0)'
> > > # Looks like you failed 1 test of 1.
> > > t/apache/singleton/process/printer_instance.t .. 
> > > Dubious, test returned 1 (wstat 256, 0x100)
> > > Failed 1/1 subtests 
> > > t/apache/singleton/request/printer_instance.t .. ok
> 
> Hm, builds fine here (amd64 and i386 sid cowbuilder chroot) in
> several runs.

I can reproduce this, although rather unreliably, by running the
test suite in a loop. It happens with both apache2-mpm-prefork and
apache2-mpm-worker, although it seems rarer with -worker.

The test suite uses the default Apache::TestConfig setting of MaxClients
2, but (at least) this particular test clearly expects that the requests
are served by the same process. When this doesn't happen, the above
failure results.

Patch to set MaxClients to 1 attached, I haven't seen a single failure
with this. Not sure if this should apply to Apache 1 as well; that's
not an issue for Debian though.
-- 
Niko Tyni   ntyni at debian.org
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Only-use-one-Apache-child-process-for-testing.patch
Type: text/x-diff
Size: 988 bytes
Desc: not available
URL: <http://lists.alioth.debian.org/pipermail/pkg-perl-maintainers/attachments/20120401/36ed8e05/attachment.patch>


More information about the pkg-perl-maintainers mailing list