On Aug 2, 2006, at 6:37 PM, Eric Wilhelm wrote:
Ken, is this possibly a CPAN on redhat problem?
As far as I know CPAN should have nothing to do with it. Unless
maybe it adds a few @INC entries, but I don't think it would do
anything pathological.
-Ken
Maybe that is not the test that was failing. Do all of the tests pass?
./Build test
Output is:
[EMAIL PROTECTED] mb]# ./Build test
t/basic...ok
t/compat..ok
t/destinationsok
8/113 skipped: various reasons
t/ext.ok
t/extend..ok
t/files..
On 8/2/06, Eric Wilhelm <[EMAIL PROTECTED]> wrote:
# from Jay Truesdale
# on Wednesday 02 August 2006 03:51 pm:
>perl Build test --test_files t/compat.t --verbose 1
>
>There were no errors reported and at the end:
>
>ok
>All tests successful.
>Files=1, Tests=60, 296 wallclock secs (144.91 cusr +
# from Jay Truesdale
# on Wednesday 02 August 2006 03:51 pm:
>perl Build test --test_files t/compat.t --verbose 1
>
>There were no errors reported and at the end:
>
>ok
>All tests successful.
>Files=1, Tests=60, 296 wallclock secs (144.91 cusr + 151.25 csys =
> 296.16 CPU)
Maybe that is not the t
> ---
> // include/linux/binfmts.h
> /*
> * MAX_ARG_PAGES defines the number of pages allocated for arguments
> * and envelope for the new program. 32 should suffice, this gives
> * a maximum env+arg of 128kB w/4KB pages!
> */
> #define MAX_ARG_PAGES 32
> ---
Yes, this is how it is defined on
# from Jay Truesdale
# on Wednesday 02 August 2006 09:14 am:
>> > #define MAX_ARG_PAGES 32
>> > ---
>>
>> Yes, this is how it is defined on Centos.
>
>I just checked on Suse 10.1 where I can compile Module::Build just
>fine and the #define is the same there as well!
It's the same almost everywher
# from Jay Truesdale
# on Wednesday 02 August 2006 08:54 am:
>The @INC entries are 1176 bytes from this command:
> perl -le 'print for @INC'|wc -m
The important bits should be:
perl -e 'print length join " ", @INC'
perl -e 'print length join " ", %ENV'
perl -e 'print length join " ", m
On 7/31/06, Eric Wilhelm <[EMAIL PROTECTED]> wrote:
# from Jay Truesdale
# on Monday 31 July 2006 11:54 am:
>On Suse I can build M::B just fine. I was going to try and update bash
>on Centos to a later version and see if the problem was solved, but I
>have not gotten back to that.
It's not a ba
# from Jay Truesdale
# on Monday 31 July 2006 11:54 am:
>On Suse I can build M::B just fine. I was going to try and update bash
>on Centos to a later version and see if the problem was solved, but I
>have not gotten back to that.
It's not a bash problem. It's a kernel issue. More specifically,
From: Eric Wilhelm <[EMAIL PROTECTED]>
To: "Jay Truesdale" <[EMAIL PROTECTED]>
Date: Sat, 29 Jul 2006 19:33:33 -0700
Subject: Fwd: Re: [Module::Build] Can't Build Module::Build
I'm not sure if Jay got on the right list. Forwarding just in case.
(Aside: I thin
# from Ken Williams
# on Saturday 29 July 2006 08:14 pm:
>>Anyone know of other stale references to that?)
>A Google search turned up only the Gmane archive:
>
>http://gmane.org/info.php?group=gmane.comp.lang.perl.modules.module-build
>
>Not sure how to fix that.
Hmm. I tried the edit link. It
On Jul 29, 2006, at 9:33 PM, Eric Wilhelm wrote:
(Aside: I think I saw Ken fix the reference to the old list in a
recent
commit. Anyone know of other stale references to that?)
A Google search turned up only the Gmane archive:
http://gmane.org/info.php?group=gmane.comp.lang.perl.modules
edFoo system if this is still an open issue.
--Eric
-- Forwarded Message: --
Subject: Re: [Module::Build] Can't Build Module::Build
Date: Saturday 29 July 2006 10:22 am
From: Chris Dolan <[EMAIL PROTECTED]>
To: Ken Williams <[EMAIL PROTECTED]>
Cc: List - Module-B
On Jul 28, 2006, at 7:23 PM, Ken Williams wrote:
Hi Jay,
The error you're getting from cpanplus must be cpanplus's fault -
as you'll see if you unpack the Module-Build tarball manually,
there's no _build/ directory or Build file present. It must be
trying to check the signature after per
Hi Jay,
The error you're getting from cpanplus must be cpanplus's fault - as
you'll see if you unpack the Module-Build tarball manually, there's
no _build/ directory or Build file present. It must be trying to
check the signature after performing some of the build steps.
The "Argument li
Hi Jay,
Note: the mailing list has moved to module-build@perl.org
# from Jay Truesdale
# on Thursday 27 July 2006 10:40 am:
>Couldn't run Build.PL: Argument list too long
I think we've seen this on redhat before.
The full test output might be helpful
cpan> look Module::Build
$ perl Build.
16 matches
Mail list logo
