develooper Front page | perl.perl5.porters | Postings from August 2001

RE: Report /pro/3gl/CPAN/perl-current@11726

Thread Previous | Thread Next
From:
Paul Marquess
Date:
August 28, 2001 08:38
Subject:
RE: Report /pro/3gl/CPAN/perl-current@11726
Message ID:
AIEAJICLCBDNAAOLLOKLOENCCNAA.Paul_Marquess@Yahoo.co.uk
From: H.Merijn Brand [mailto:h.m.brand@hccnet.nl]

> On Fri 24 Aug 2001 10:36, "Paul Marquess"
> <Paul_Marquess@Yahoo.co.uk> wrote:
> > From: H.Merijn Brand [mailto:h.m.brand@hccnet.nl]
...
>
> Great, that'll make a lot of smoke reports happy (p5p does *not*
> see them all,
> Schwern and I filter what we think is useful)
>
> > I'm not clear on how to read the output from the smoke test. Is DB_File
> > failing in *all* the configuration combinations for AIX 4.3?
> Also, is test
>
> No, here's this morning's version (still running) with some comments
>
> --8<---
> Automated smoke report for patch 11735
>                                                       |
>   cc           gcc
> O = OK                                                | HP-UX
> 11.00 B.11.11.02   3.0.1           32-bit
> F = Failure(s), extended report at the bottom         |
>                2.9-hppa-000310 64-bit
> ? = still running or test results not (yet) available | HP-UX
> 10.20 A.10.32.30   2.8.1
> Build failures during:       - = unknown,   = skipped | AIX
> 4.3.3.0 vac 5.0.2.0  3.0.1
>     c = Configure, m = make, t = make test-prep       | AIX
> 4.2.1.0 xlc 3.1.4.10 2.8.1
> -->8---
>
> For these systems using these compilers,              For these
> configurations
> --8<---
>  HP-UX    HP-UX    HP-UX     AIX      AIX      AIX
>  11.00    11.00    10.20     4.3      4.3      4.2
>   HPc      gcc      HPc      vac      gcc      xlc
>  11735    11735    11735    11735    11726    11735   Configuration
> -------  -------  -------  -------  -------  -------
> --------------------------------------------------------------------
> O O O O  O O O O  m m m m  c c c c  O O O O  O O O O  -Uuseperlio
> O O O O  O O O O  m m m m  c c c c  O O O O  O O O O  -Duseperlio
> O O O O  F F F F  c c c c  c c c c  c c c c  c c c c  -Duseperlio
> -Duse64bitint
> O O O O  F F F F  c c c c  c c t t           c c c c  -Duseperlio
> -Duse64bitall
> O O O O  c c c c  m m m m  O O t t  O O O O  O O O O  -Duseperlio
> -Duselongdouble
> O O O O  c c c c  c c c c  O O t t           c c c c  -Duseperlio
> -Dusemorebits
> O O O O  F F F F  c c c c  O O t t  c c c c  c c c c  -Duseperlio
> -Duse64bitall -Duselongdouble
> O O O O  O O O O  t t F F  O O O O  O O O O  O O O O  -Uuseperlio
> -Dusethreads -Duseithreads
> O O O O  O O O ?  F F F F  O O O O  O O c c  O O O O  -Duseperlio
> -Dusethreads -Duseithreads
> O O O O  ? ? ? ?  c c c c  F F F F  c c c c  c c c c  -Duseperlio
> -Duse64bitint -Dusethreads -Duseithreads
> O O O O  ? ? ? ?  c c c c  F F F F           c c c c  -Duseperlio
> -Duse64bitall -Dusethreads -Duseithreads
> O O O O  ? ? ? ?  F F F F  O O O O  c c c c  O ? ? ?  -Duseperlio
> -Duselongdouble -Dusethreads -Duseithreads
> O O O O  ? ? ? ?  c c c c  F F ? ?           ? ? ? ?  -Duseperlio
> -Dusemorebits -Dusethreads -Duseithreads
> O O O O  ? ? ? ?  c c c c  ? ? ? ?  c c c c  ? ? ? ?  -Duseperlio
> -Duse64bitall -Duselongdouble -Dusethreads -Duseithreads
> | |                            | |
> | +----- PERLIO = perlio       | +- PERLIO = perlio -DDEBUGGING
> +------- PERLIO = stdio        +--- PERLIO = stdio  -DDEBUGGING
> -->8---
> Each configuration consists of 4 letters: 'O O O O' means all
> test passed for
> PERLIO=stdio without -DDEBUGGIN, PERLIO=perlio without -DDEBUGGING,
> PERLIO=stdio with -DDEBUGGING and PERLIO=perlio with -DDEBUGGING' in that
> order. Interesting patters are 'O F O F' (stdio OK, perlio
> fails), 'O O F F'
> (without -DDEBUGGING OK, with -DDEBUGGING fail) and like above 'c
> c t t' which
> needs some invesigation (probably disk full or alike, because the
> 'O's start
> appearing later and the 'O O t t' is a week or so old and already
> posted to
> Robin, because it has to do with ext/B
>
> For every 'F' in above list, I'll show the errors that were
> detected in that
> configuration, but only the first for each test to save time and
> (report) space
> (OTOH, this might change in future smokes)
> --8<---
>
> Failures:
>
> HPUX 11 gcc  stdio/perlio     -Duseperlio -Duse64bitint
> HPUX 11 gcc  stdio/perlio     -DDEBUGGING -Duseperlio -Duse64bitint
> HPUX 11 gcc  stdio/perlio     -Duseperlio -Duse64bitall
> HPUX 11 gcc  stdio/perlio     -DDEBUGGING -Duseperlio -Duse64bitall
> HPUX 11 gcc  stdio/perlio     -Duseperlio -Duse64bitall -Duselongdouble
> HPUX 11 gcc  stdio/perlio     -DDEBUGGING -Duseperlio
> -Duse64bitall -Duselongdouble
>     ext/IO/lib/IO/t/io_multihomed......FAILED at test 3
>     lib/Net/hostent....................FAILED at test 4
>
> HP-UX 10.20  stdio/perlio     -DDEBUGGING -Uuseperlio
> -Dusethreads -Duseithreads
> HP-UX 10.20  stdio/perlio     -Duseperlio -Dusethreads -Duseithreads
> HP-UX 10.20  stdio/perlio     -DDEBUGGING -Duseperlio
> -Dusethreads -Duseithreads
> HP-UX 10.20  stdio/perlio     -Duseperlio -Duselongdouble
> -Dusethreads -Duseithreads
> HP-UX 10.20  stdio/perlio     -DDEBUGGING -Duseperlio
> -Duselongdouble -Dusethreads -Duseithreads
>     lib/File/Find/taint................FAILED at test 29
>
> AIX 4.3      stdio/perlio     -Duseperlio -Duse64bitint
> -Dusethreads -Duseithreads
> AIX 4.3      stdio/perlio     -DDEBUGGING -Duseperlio
> -Duse64bitint -Dusethreads -Duseithreads
> AIX 4.3      stdio/perlio     -Duseperlio -Duse64bitall
> -Dusethreads -Duseithreads
> AIX 4.3      stdio/perlio     -DDEBUGGING -Duseperlio
> -Duse64bitall -Dusethreads -Duseithreads
> AIX 4.3      stdio/perlio     -Duseperlio -Dusemorebits
> -Dusethreads -Duseithreads
>     ext/DB_File/t/db-btree.............FAILED at test 79
> -->8---
>
> You should be able to find a line for every 'F' in the matrix.
>
> > 79 the only test that fails or does the smoke test only report the first
> > one?
>
> The smoke only reports the first failure. If I'm lkucky, it *did*
> store the
> harness output in a log. Looking ...
>
> No, that needs examining ;-P
>
> Running it from the current smoke situation (it's now smoking
> configuration
> -Dusedevel -Duseperlio -Dusemorebits -Dusethreads -Duseithreads), I get
>
> i2:/pro/3gl/CPAN/perl-current/t 160 > ./perl harness
> ../ext/DB_File/t/db-btree.t
> ../ext/DB_File/t/db-btree....FAILED tests 79, 82, 85, 150-153
>         Failed 7/157 tests, 95.54% okay
> Failed Test                 Stat Wstat Total Fail  Failed  List of Failed
> ------------------------------------------------------------------
> -------------
> ../ext/DB_File/t/db-btree.t              157    7   4.46%  79 82
> 85 150-153
> Failed 1/1 test scripts, 0.00% okay. 7/157 subtests failed, 95.54% okay.
> i2:/pro/3gl/CPAN/perl-current/t 161 >
>
> running as plain './perl ../ext/etc... didn't produce any more
> useful info.

Aha! There is a distinct pattern to those failures. All are involved in
testing the Berkeley DB feature where you can have duplicate keys.

> > Oh yea, what version of Berkeley DB is running on the AIX box.
> There are a
> > few versions of Berkeley DB kicking about that have problems. Here is a
> > one-liner that should tell you.
> >
> >     perl -e 'use DB_File; print "Berkeley DB ver $DB_File::db_ver\n"'
>
> Berkeley DB ver 3.002009
>
> I know 3.3.11 is available (maybe even newer), but I have not
> found time to
> install yet.

I'm not aware of any problems with 3.2.9, but you never know.

Could you try running this script on one of the failing AIX configurations
please?

use strict ;
use DB_File ;

sub Put
{
    my $obj = shift ;
    my $key = shift ;
    my $value = shift ;

    my $status = $obj->put($key, $value) ;
    print "Put [$key] [$value], status [$status]\n" ;
}

my $dup = R_DUP ;

print <<EOM ;
This is Perl        $]
Running DB_File     $DB_File::VERSION
Running Berkeley DB $DB_File::db_ver
R_DUP               $dup

EOM

my $filename = "tree.$$" ;
my %h ;
unlink $filename ;

# Enable duplicate records
$DB_BTREE->{'flags'} = R_DUP ;

my $H = tie %h, "DB_File", $filename, O_RDWR|O_CREAT, 0640, $DB_BTREE
    or die "Cannot open $filename: $!\n";

# Add some key/value pairs to the file
Put($H, 'Wall', 'Larry') ;
Put($H, 'Wall', 'Brick') ;
Put($H, 'Wall', 'Brick') ;
Put($H, 'Smith', 'John') ;
Put($H, 'mouse', 'mickey') ;

my $key  = 0 ;
my $value = 0 ;
my $status ;
for ($status = $H->seq($key, $value, R_FIRST) ;
     $status == 0 ;
     $status = $H->seq($key, $value, R_NEXT) )
  {  print "$key    -> $value\n" }

print "Done - status = [$status]\n" ;


untie %h ;


Paul


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


Thread Previous | Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About