Bug#665221: libdancer-plugin-database-perl: FTBFS: tests failed

2012-06-10 Thread Niko Tyni
On Sat, Jun 09, 2012 at 07:15:21PM +0200, intrigeri wrote:

 I'd like the RC part of the problem to be fixed in time for Wheezy,
 so I suggest we:
 
  1. patch the failing test to become a TODO one
  2. add Niko's TODO minimal test case (to be notified if the
 bug is fixed upstream)
  3. file a new (non-RC) Debian bug about the actual bug that's
 demonstrated by the FTBFS
  4. close this FTBFS bug

OK by me (but you could also just downgrade the existing bug after adding
the workaround.)

It would be good to ping upstream on a mailing list or something,
I don't know how actively they use the github issue tracker.

Thanks for looking at this,
-- 
Niko Tyni   nt...@debian.org



-- 
To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org



Bug#665221: libdancer-plugin-database-perl: FTBFS: tests failed

2012-06-09 Thread intrigeri
Hi,

Niko Tyni wrote (31 Mar 2012 07:46:37 GMT) :
 I'm attaching a test script that should demonstrate this
 in a reproducible way.

I'm impressed by the detailed debugging you did!
Congrats :)

 I've forwarded this upstream as 
  https://github.com/bigpresh/Dancer-Plugin-Database/issues/29.

... unfortunately, they did not comment yet.

I'd like the RC part of the problem to be fixed in time for Wheezy,
so I suggest we:

 1. patch the failing test to become a TODO one
 2. add Niko's TODO minimal test case (to be notified if the
bug is fixed upstream)
 3. file a new (non-RC) Debian bug about the actual bug that's
demonstrated by the FTBFS
 4. close this FTBFS bug

Niko, what do you think?

Cheers,
--
  intrigeri
  | GnuPG key @ https://gaffer.ptitcanardnoir.org/intrigeri/intrigeri.asc
  | OTR fingerprint @ https://gaffer.ptitcanardnoir.org/intrigeri/otr.asc



-- 
To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org



Processed: Re: Bug#665221: libdancer-plugin-database-perl: FTBFS: tests failed

2012-03-31 Thread Debian Bug Tracking System
Processing commands for cont...@bugs.debian.org:

 tag 665221 - moreinfo unreproducible
Bug #665221 [src:libdancer-plugin-database-perl] 
libdancer-plugin-database-perl: FTBFS: tests failed
Removed tag(s) unreproducible and moreinfo.
 tag 665221 + upstream
Bug #665221 [src:libdancer-plugin-database-perl] 
libdancer-plugin-database-perl: FTBFS: tests failed
Added tag(s) upstream.
 forwarded 665221 https://github.com/bigpresh/Dancer-Plugin-Database/issues/29
Bug #665221 [src:libdancer-plugin-database-perl] 
libdancer-plugin-database-perl: FTBFS: tests failed
Set Bug forwarded-to-address to 
'https://github.com/bigpresh/Dancer-Plugin-Database/issues/29'.
 thanks
Stopping processing here.

Please contact me if you need assistance.
-- 
665221: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=665221
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems


--
To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org



Bug#665221: libdancer-plugin-database-perl: FTBFS: tests failed

2012-03-31 Thread Niko Tyni
tag 665221 - moreinfo unreproducible
tag 665221 + upstream
forwarded 665221 https://github.com/bigpresh/Dancer-Plugin-Database/issues/29
thanks

On Thu, Mar 22, 2012 at 05:34:55PM +0100, Lucas Nussbaum wrote:
 Source: libdancer-plugin-database-perl
 Version: 1.81-1
 Severity: serious
 Tags: wheezy sid
 User: debian...@lists.debian.org
 Usertags: qa-ftbfs-20120321 qa-ftbfs
 Justification: FTBFS on amd64

  #   Failed test 'database_connection_failed hook fires'
  #   at t/01-basic.t line 155.
  #  got: ''
  # expected: '1'
  # Looks like you failed 1 test of 41.

This one is quite a heisenbug. Fortunately it's reproducible
here with 'make test' (but not with 'prove -b t/*.t').

The problem is in using of argument hashref stringification as
a database handle index in Dancer::Plugin::Database::database().
Quoting some comments in the code:

  # Hashref used as key for default handle, so we don't have a magic value that
  # the user could use for one of their connection names and cause problems
  # (Kudos to Igor Bujna for the idea)
  [...]
# The key to use to store this handle in %handles.  This will be either the
# name supplied to database(), the hashref supplied to database() (thus, as
# long as the same hashref of settings is passed, the same handle will be
# reused) or $def_handle if database() is called without args:
  [...]
# Accept a hashref of settings to use, if desired.  If so, we use this
# hashref to look for the handle, too, so as long as the same hashref is
# passed to the database() keyword, we'll reuse the same handle:

This fails when memory is reused so that a different argument hash gets
the same memory address and hence the same stringification.

Adding this instrumentation into Database.pm:60 or thereabouts:

 use Data::Dumper;
 my $new = Dumper($handle_key);

 if (exists $strings{$handle_key}) {
 my $old = $strings{$handle_key};
 warn hashref collision: $old vs. $new if $old ne $new;
 }
 $strings{$handle_key} = $new;

gives this output when the test fails:

   t/01-basic.t .. 40/41 hashref collision: $VAR1 = {
 'database' = ':memory:',
 'driver' = 'SQLite'
   };
vs. $VAR1 = {
 'dsn' = 'dbi:SQLite:/Please/Tell/Me/This/File/Does/Not/Exist!',
 'dbi_params' = {
   'HandleError' = sub { DUMMY },
   'RaiseError' = 0,
   'PrintError' = 0
 }
   };
   
   #   Failed test 'database_connection_failed hook fires'
   #   at t/01-basic.t line 155.
   #  got: ''
   # expected: '1'
   # Looks like you failed 1 test of 41.
   
which clearly shows the problem: the test is making sure that a
nonexistent database can't be opened, but the code is reusing a handle
for an in-memory database that's working fine.

I'm attaching a test script that should demonstrate this
in a reproducible way.

I suppose the handle index should include a serialization of the
arguments.  However, I'm afraid that's hard to do for the general case
as they may include code references (like the 'HandleError' subroutine
above).

I've forwarded this upstream as 
 https://github.com/bigpresh/Dancer-Plugin-Database/issues/29.
-- 
Niko Tyni   nt...@debian.org


03-serial.t
Description: Troff document


Bug#665221: libdancer-plugin-database-perl: FTBFS: tests failed

2012-03-23 Thread Salvatore Bonaccorso
tags 665221 + moreinfo unreproducible
thanks

Hi

On Thu, Mar 22, 2012 at 05:34:55PM +0100, Lucas Nussbaum wrote:
 Source: libdancer-plugin-database-perl
 Version: 1.81-1
 Severity: serious
 Tags: wheezy sid
 User: debian...@lists.debian.org
 Usertags: qa-ftbfs-20120321 qa-ftbfs
 Justification: FTBFS on amd64
 
 Hi,
 
 During a rebuild of all packages in sid, your package failed to build on
 amd64.
 
 Relevant part:
  make[1]: Entering directory `/«PKGBUILDDIR»'
  PERL_DL_NONLAZY=1 /usr/bin/perl -MExtUtils::Command::MM -e 
  test_harness(0, 'blib/lib', 'blib/arch') t/*.t
  # Testing Dancer::Plugin::Database 1.81, Perl 5.014002, /usr/bin/perl
  t/00-load.t ... ok
  
  #   Failed test 'database_connection_failed hook fires'
  #   at t/01-basic.t line 155.
  #  got: ''
  # expected: '1'
  # Looks like you failed 1 test of 41.
  t/01-basic.t .. 
  Dubious, test returned 1 (wstat 256, 0x100)
  Failed 1/41 subtests 
  # Testing Dancer::Plugin::Database::Handle 0.12, Perl 5.014002, 
  /usr/bin/perl
  t/02-handle.t . ok
  t/manifest.t .. skipped: Author tests not required for installation
  t/pod-coverage.t .. ok
  t/pod.t ... ok
  
  Test Summary Report
  ---
  t/01-basic.t(Wstat: 256 Tests: 41 Failed: 1)
Failed test:  41
Non-zero exit status: 1
  Files=6, Tests=50,  2 wallclock secs ( 0.07 usr  0.02 sys +  0.68 cusr  
  0.11 csys =  0.88 CPU)
  Result: FAIL
  Failed 1/6 test programs. 1/50 subtests failed.
  make[1]: *** [test_dynamic] Error 255

I was not able to reproduce this (within an sbuild environment and too
on a VM). Someone has an idea?

Regards,
Salvatore


signature.asc
Description: Digital signature


Processed: Re: Bug#665221: libdancer-plugin-database-perl: FTBFS: tests failed

2012-03-23 Thread Debian Bug Tracking System
Processing commands for cont...@bugs.debian.org:

 tags 665221 + moreinfo unreproducible
Bug #665221 [src:libdancer-plugin-database-perl] 
libdancer-plugin-database-perl: FTBFS: tests failed
Added tag(s) moreinfo.
 thanks
Stopping processing here.

Please contact me if you need assistance.
-- 
665221: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=665221
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems


--
To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org



Bug#665221: libdancer-plugin-database-perl: FTBFS: tests failed

2012-03-23 Thread gregor herrmann
On Fri, 23 Mar 2012 13:09:27 +0100, Salvatore Bonaccorso wrote:

 I was not able to reproduce this (within an sbuild environment and too
 on a VM). Someone has an idea?

I also can't reproduce it.
(Sorry for quietly tagging it without a comment.)

Cheers,
gregor

-- 
 .''`.  Homepage: http://info.comodo.priv.at/ - OpenPGP key 0xBB3A68018649AA06
 : :' : Debian GNU/Linux user, admin, and developer  -  http://www.debian.org/
 `. `'  Member of VIBE!AT  SPI, fellow of the Free Software Foundation Europe
   `-   


signature.asc
Description: Digital signature


Bug#665221: libdancer-plugin-database-perl: FTBFS: tests failed

2012-03-22 Thread Lucas Nussbaum
Source: libdancer-plugin-database-perl
Version: 1.81-1
Severity: serious
Tags: wheezy sid
User: debian...@lists.debian.org
Usertags: qa-ftbfs-20120321 qa-ftbfs
Justification: FTBFS on amd64

Hi,

During a rebuild of all packages in sid, your package failed to build on
amd64.

Relevant part:
 make[1]: Entering directory `/«PKGBUILDDIR»'
 PERL_DL_NONLAZY=1 /usr/bin/perl -MExtUtils::Command::MM -e 
 test_harness(0, 'blib/lib', 'blib/arch') t/*.t
 # Testing Dancer::Plugin::Database 1.81, Perl 5.014002, /usr/bin/perl
 t/00-load.t ... ok
 
 #   Failed test 'database_connection_failed hook fires'
 #   at t/01-basic.t line 155.
 #  got: ''
 # expected: '1'
 # Looks like you failed 1 test of 41.
 t/01-basic.t .. 
 Dubious, test returned 1 (wstat 256, 0x100)
 Failed 1/41 subtests 
 # Testing Dancer::Plugin::Database::Handle 0.12, Perl 5.014002, /usr/bin/perl
 t/02-handle.t . ok
 t/manifest.t .. skipped: Author tests not required for installation
 t/pod-coverage.t .. ok
 t/pod.t ... ok
 
 Test Summary Report
 ---
 t/01-basic.t(Wstat: 256 Tests: 41 Failed: 1)
   Failed test:  41
   Non-zero exit status: 1
 Files=6, Tests=50,  2 wallclock secs ( 0.07 usr  0.02 sys +  0.68 cusr  0.11 
 csys =  0.88 CPU)
 Result: FAIL
 Failed 1/6 test programs. 1/50 subtests failed.
 make[1]: *** [test_dynamic] Error 255

The full build log is available from:
   
http://people.debian.org/~lucas/logs/2012/03/21/libdancer-plugin-database-perl_1.81-1.log

A list of current common problems and possible solutions is available at 
http://wiki.debian.org/qa.debian.org/FTBFS . You're welcome to contribute!

About the archive rebuild: The rebuild was done on about 50 AMD64 nodes
of the Grid'5000 platform, using a clean chroot.  Internet was not
accessible from the build systems.



-- 
To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org