[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Can backports be built with backports ?




On 24 Nov 2009, at 2:20 pm, Charles Plessy wrote:

I would like to backport the bioperl package as well (a bigger beast), and it needs ExtUtils::Manifest >= 1.52 to build (but not to run). However Lenny's
perl-modules package does not provide such a high version.

Hehe, hi Charles. Now you know why, when I discussed this with you a couple of years ago at Debconf, that I said I didn't use Debian packaged sequencing packages. All the dependencies are a nightmare, the scientists always want the very latest of this, and simultaneously some ancient version of that, and they want multiple versions of everything installed simultaneously. It's a nightmare.

That's not to say that I don't think you and the Debian Med team are doing a fabulous job. I'm just saying I really don't envy you. Me, I just install all this crap on an NFS server and be done with it. Our / software central NFS server now contains more than 600GB of various versions of software the scientists want. Actually, I don't install most of it. Fortunately, they're allowed to build their own stuff, which makes my life a bit simpler (and accounts for a lot of the size - multiple installs for different groups of different versions of apache, tomcat, bioperl, ssaha, blast, you name it)

Still 600GB is nothing in the world of next-gen sequencing, n'est-ce pas? We've got a 0.5 petabyte Lustre filesystem just as a staging area for the sequencers... 600GB is less than one disk spindle.

Tim


--
The Wellcome Trust Sanger Institute is operated by Genome Research Limited, a charity registered in England with number 1021457 and a company registered in England with number 2742969, whose registered office is 215 Euston Road, London, NW1 2BE.
Reply to: