perl-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Geoffrey Young <>
Subject Re: AT Bug
Date Wed, 16 Jul 2008 13:23:03 GMT

Torsten Foertsch wrote:
> On Wed 16 Jul 2008, Geoffrey Young wrote:
>> I'd be interested in
>> recording all the values passed in and see if they match during start
>> and restart.
> here you are.
> The attached file is written with this print statement:
>       print $f "$$=".(syscall 39).": ".($path // "(UNDEF)").": ".
> 	join( ", ", map {my @l=caller $_; "[$l[1]($l[2]): $l[3]]"} 0..3 )."\n";
> I added syscall(39) (getpid) to make sure the correct $$ is reported.

yeah, so the paths are undef the second time around.  if I had to guess, 
I'd say that taint is unsetting PERL5LIB or PATH or something and that 
yes, the error_log the second time around is /dev/null.  I doubt there 
is anything we can do to "fix" the "problem" since both portions are out 
of our hands and are probably appropriate.

making Apache::Test not blow up seems like a good compromise :)


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message