Re: Perl 5.12.2-RC1 is now available

All,

it's getting weirder and weirder. Again I threw away everything
(MMK REALCLEAN), re-logged-in again, cleaned up symbols (apart
from MMK), looked carefully at logicals, removed any DECC
feature logicals, and started a new build - which failed at
a different location with a different error... <sigh>

Please have a look at the attached log file. I tried to list
everything relevant to the build environment.

I will re-unpack the distribution to really have a clean start.

cu,
Martin

-----Ursprüngliche Nachricht-----
Von: Craig A. Berry [mailto:[email protected]] 
Gesendet: Donnerstag, 2. September 2010 02:19
An: [email protected]; Vorländer, Martin
Cc: VMSperl Mailing List
Betreff: Re: Perl 5.12.2-RC1 is now available


On Sep 1, 2010, at 9:15 AM, Vorländer, Martin wrote:

> MUCH better:
>
> Failed 37 tests out of 1611, 97.70% okay.

Thanks for the report.  I downloaded the tarball and built a default  
configuration and got:

All tests successful.
u=67.65  s=0.00  cu=0.00  cs=0.00  scripts=1631  tests=327531

Note that I ran 1631 test scripts and you only ran 1611, so there must  
be some feature (possibly symlinks?) that I have but you don't.

>      ../cpan/Archive-Extract/t/01_Archive-Extract.t
>        ../cpan/CPANPLUS-Dist-Build/t/02_CPANPLUS-Dist-Build.t
<snip>
>        ../cpan/ExtUtils-MakeMaker/t/basic.t
<snip>
>        ../cpan/Module-Build/t/PL_files.t
<snip>
>        ../cpan/Time-HiRes/t/HiRes.t
>        ../dist/ExtUtils-Install/t/Install.t
<snip>

All of these groups probably fail as a group, and probably because  
they aren't finding something they want.  Sometimes a PATH definition  
can be the problem.  Hard to say without details.

>        ../lib/perl5db.t

This is testing the debugger.  It might get confused by a logical name  
like TT, or TTY or TERM or something.

>        ../lib/vmsish.t

This tends to have trouble for people on the plus side of UTC time.

>        op/stat.t
>        porting/manifest.t

No idea.

> u=53.58  s=0.00  cu=0.00  cs=0.00  scripts=1611  tests=316018
>
> Anything I should look into immediately?

If you have time to look into the vmsish.t failure, that might be  
something we should sort out.

________________________________________
Craig A. Berry
mailto:[email protected]

"... getting out of a sonnet is much more
  difficult than getting in."
                  Brad Leithauser