I'll post 0.58001 tonight, which has tested for mm-config
as you suggest, for quite some (unpublished) time. I was
waiting so that testers would have mm loaded, but there's
no right way to stage such events.
The "appears to hang" may be the result of using /dev/random
before it has enough randomness. 0.58001 uses /dev/urandom
to avoid such hangups. It seems that Linux takes
/dev/randomness more seriously than FreeBSD which I use.
Thanks,
Craig MacKenna
On Nov 4, 2009, at 4:26 PM, David Cantrell wrote:
cr...@animalhead.com wrote:
The mm library, which is a non-CPAN prerequisite for IPC::MMA, is
quite
easy to download and install from http://www.ossp.org/pkg/lib/mm/
$ sudo apt-get install libmm-dev
...
$ perl-5.6.2/bin/cpan IPC::MMA
...
Please enter path to libmm.so: [/usr/local/lib]
that's the wrong path (should be /usr/lib). I see that later in
Makefile.PL you run 'which mm-config' to see if you can find that,
so I
suggest doing that earlier, and if you can find it, interrogate it for
the path to use.
The path is even wronger on my Solaris testing box - there it's
/usr/local/home/david/cpantesting/lib. But mm-config is still in the
path, cos /usr/local/home/david/cpantesting/bin is in the path.
Anyway, with the right path, on Linux:
* 5.6.2: fail, perl appeared to hang at the end of the tests
(or possibly while generating the test report)
* 5.8.9: t/5_hash appears to hang. aborted.
* 5.10.1: t/5_hash appears to hang. aborted.
Test results for Solaris and Irix may (or may not, if they hang
too) be
forthcoming! If they don't show up on the CPAN-testers website in the
next few days, assume it didn't work.
--
David Cantrell | Cake Smuggler Extraordinaire
fdisk format reinstall, doo-dah, doo-dah;
fdisk format reinstall, it's the Windows way
