Bug #56485 for Dist-Zilla-Plugin-Git: test failure t/4-push-multi.t t/4-push.t

Sun Apr 11 11:16:58 2010 daxim [...] cpan.org - Ticket created

Subject:

test failure t/4-push-multi.t t/4-push.t

~/.cpan/build/Dist-Zilla-Plugin-Git-1.100970> prove -bv t/4* t/4-push-multi.t .. 1..6 Initialized empty Git repository in /home/daxim/.cpan/build/Dist-Zilla- Plugin-Git-1.100970-IUOefu/tmp/CzhZTxO8Fv/source/.git/ fatal: destination directory '/tmp/8esEYkoDyR' already exists. # Looks like your test exited with 128 before it could output anything. Dubious, test returned 128 (wstat 32768, 0x8000) Failed 6/6 subtests t/4-push.t ........ 1..3 Initialized empty Git repository in /home/daxim/.cpan/build/Dist-Zilla- Plugin-Git-1.100970-IUOefu/tmp/_ah3895SB7/source/.git/ fatal: destination directory '/tmp/W9QfnLMkEd' already exists. # Looks like your test exited with 128 before it could output anything. Dubious, test returned 128 (wstat 32768, 0x8000) Failed 3/3 subtests Test Summary Report ------------------- t/4-push-multi.t (Wstat: 32768 Tests: 0 Failed: 0) Non-zero exit status: 128 Parse errors: Bad plan. You planned 6 tests but ran 0. t/4-push.t (Wstat: 32768 Tests: 0 Failed: 0) Non-zero exit status: 128 Parse errors: Bad plan. You planned 3 tests but ran 0. Files=2, Tests=0, 4 wallclock secs ( 0.04 usr 0.01 sys + 2.36 cusr 0.30 csys = 2.71 CPU) Result: FAIL

Mon May 03 11:05:54 2010 jquelin [...] cpan.org - Correspondence added

On Sun Apr 11 11:16:58 2010, DAXIM wrote: Show quoted text

> ~/.cpan/build/Dist-Zilla-Plugin-Git-1.100970> prove -bv t/4* > t/4-push-multi.t .. > 1..6 > Initialized empty Git repository in /home/daxim/.cpan/build/Dist-Zilla- > Plugin-Git-1.100970-IUOefu/tmp/CzhZTxO8Fv/source/.git/ > fatal: destination directory '/tmp/8esEYkoDyR' already exists. > # Looks like your test exited with 128 before it could output anything. > Dubious, test returned 128 (wstat 32768, 0x8000) > Failed 6/6 subtests

this is weird... i'm using file::temp::tempdir() to create a new, unique (supposedly) directory. can you reproduce this behaviour at will?

Mon May 03 11:05:55 2010 The RT System itself - Status changed from 'new' to 'open'

Mon May 03 11:21:51 2010 daxim [...] cpan.org - Correspondence added

No, I can't reproduce it anymore. Marking this a reject.

Mon May 03 11:21:52 2010 daxim [...] cpan.org - Status changed from 'open' to 'rejected'

Mon May 03 11:30:42 2010 jquelin [...] cpan.org - Correspondence added

On Mon May 03 11:21:51 2010, DAXIM wrote: Show quoted text

> No, I can't reproduce it anymore. Marking this a reject.

ok, thanks. reopen if you see the error popping once again...

Mon May 03 11:30:44 2010 The RT System itself - Status changed from 'rejected' to 'open'

Sat May 08 08:11:01 2010 polettix [...] cpan.org - Correspondence added

Il Lun 03 Maggio 2010 11:30:42, JQUELIN ha scritto: Show quoted text

> On Mon May 03 11:21:51 2010, DAXIM wrote:

> > No, I can't reproduce it anymore. Marking this a reject.

> > ok, thanks. > reopen if you see the error popping once again...

Yes I can, this happens all the time: /opt/perl/.cpan/build/Dist-Zilla-Plugin-Git-1.101230-EPJzsD$ prove -bv t/4* t/4-push-multi.t .. 1..6 Initialized empty Git repository in .git/ destination directory '/tmp/FYiD319qRl' already exists. # Looks like your test exited with 1 before it could output anything. Dubious, test returned 1 (wstat 256, 0x100) Failed 6/6 subtests t/4-push.t ........ 1..3 Initialized empty Git repository in .git/ destination directory '/tmp/eFiCzWZu4X' already exists. # Looks like your test exited with 1 before it could output anything. Dubious, test returned 1 (wstat 256, 0x100) Failed 3/3 subtests Test Summary Report ------------------- t/4-push-multi.t (Wstat: 256 Tests: 0 Failed: 0) Non-zero exit status: 1 Parse errors: Bad plan. You planned 6 tests but ran 0. t/4-push.t (Wstat: 256 Tests: 0 Failed: 0) Non-zero exit status: 1 Parse errors: Bad plan. You planned 3 tests but ran 0. Files=2, Tests=0, 2 wallclock secs ( 0.02 usr 0.01 sys + 1.44 cusr 0.46 csys = 1.93 CPU) Result: FAIL /opt/perl/.cpan/build/Dist-Zilla-Plugin-Git-1.101230-EPJzsD$ prove -bv t/4* t/4-push-multi.t .. 1..6 Initialized empty Git repository in .git/ destination directory '/tmp/sDk2HZHIlf' already exists. # Looks like your test exited with 1 before it could output anything. Dubious, test returned 1 (wstat 256, 0x100) Failed 6/6 subtests t/4-push.t ........ 1..3 Initialized empty Git repository in .git/ destination directory '/tmp/89xuY6IAFJ' already exists. # Looks like your test exited with 1 before it could output anything. Dubious, test returned 1 (wstat 256, 0x100) Failed 3/3 subtests Test Summary Report ------------------- t/4-push-multi.t (Wstat: 256 Tests: 0 Failed: 0) Non-zero exit status: 1 Parse errors: Bad plan. You planned 6 tests but ran 0. t/4-push.t (Wstat: 256 Tests: 0 Failed: 0) Non-zero exit status: 1 Parse errors: Bad plan. You planned 3 tests but ran 0. Files=2, Tests=0, 2 wallclock secs ( 0.02 usr 0.01 sys + 1.51 cusr 0.48 csys = 2.02 CPU) Result: FAIL

Sat May 08 08:17:24 2010 polettix [...] cpan.org - Correspondence added

Il Sab 08 Maggio 2010 08:11:01, POLETTIX ha scritto: Show quoted text

> Il Lun 03 Maggio 2010 11:30:42, JQUELIN ha scritto:

> > On Mon May 03 11:21:51 2010, DAXIM wrote:

> > > No, I can't reproduce it anymore. Marking this a reject.

> > > > ok, thanks. > > reopen if you see the error popping once again...

> > Yes I can, this happens all the time:

The offending line is the following: $git->clone( { quiet=>1, 'no-checkout'=>1, bare=>1 }, $curr, $clone ); I suspect this has to do with the fact that I'm using git 1.5.x as found in my Debian. git version is giving me other troubles in other tests, so I suspect I'm forced to install git by myself... I'll let you know how the test goes after git's upgrade.

Sat May 08 08:40:34 2010 polettix [...] cpan.org - Correspondence added

Il Sab 08 Maggio 2010 08:17:24, POLETTIX ha scritto: Show quoted text

> Il Sab 08 Maggio 2010 08:11:01, POLETTIX ha scritto:

> > Il Lun 03 Maggio 2010 11:30:42, JQUELIN ha scritto:

> > > On Mon May 03 11:21:51 2010, DAXIM wrote:

> > > > No, I can't reproduce it anymore. Marking this a reject.

> > > > > > ok, thanks. > > > reopen if you see the error popping once again...

> > > > Yes I can, this happens all the time:

> > The offending line is the following: > > $git->clone( { quiet=>1, 'no-checkout'=>1, bare=>1 }, $curr, $clone ); > > I suspect this has to do with the fact that I'm using git 1.5.x as > found in my Debian. git > version is giving me other troubles in other tests, so I suspect I'm > forced to install git by > myself... I'll let you know how the test goes after git's upgrade.

Hi there, I confirm that upgrading to git 1.7.1 solves the problem.

Sat May 08 12:17:31 2010 jquelin [...] cpan.org - Correspondence added

On Sat May 08 08:40:34 2010, POLETTIX wrote: Show quoted text

> Hi there, I confirm that upgrading to git 1.7.1 solves the problem.

hmm. so what to do now? do you think the module should barf during install if git is not recent enough? or at least wrap those tests in a skip block? i'm not convinced...

Sun Jun 20 02:33:00 2010 MARKF [...] cpan.org - Correspondence added

On Sat May 08 12:17:31 2010, JQUELIN wrote: Show quoted text

> On Sat May 08 08:40:34 2010, POLETTIX wrote:

> > Hi there, I confirm that upgrading to git 1.7.1 solves the problem.

> > hmm. so what to do now? do you think the module should barf during > install if git is not recent enough? > or at least wrap those tests in a skip block? > > i'm not convinced...

At the very least you could get the tests scripts to complain "uh oh, looks like you've got an old version of GIT, these tests might fail, try upgrading" (I got this error and the only way I worked out what the problem was was coming to RT to file a bug report and finding this thread)

Tue Jun 29 13:00:46 2010 jquelin [...] cpan.org - Correspondence added

should be fixed in Git-1.101801

Tue Jun 29 13:00:47 2010 jquelin [...] cpan.org - Status changed from 'open' to 'resolved'

Tue Jun 29 13:01:55 2010 jquelin [...] cpan.org - Correspondence added

in Dist-Zilla-Plugin-Git-1.101801, sorry

Tue Jun 29 13:01:56 2010 The RT System itself - Status changed from 'resolved' to 'open'

Tue Jun 29 13:02:32 2010 jquelin [...] cpan.org - Status changed from 'open' to 'resolved'