[155] in athena10
Re: /svn/athena r22816 - in trunk/debathena/meta: . cluster-software
daemon@ATHENA.MIT.EDU (Timothy G Abbott)
Mon Apr 7 10:57:40 2008
Date: Mon, 7 Apr 2008 10:55:03 -0400 (EDT)
From: Timothy G Abbott <tabbott@MIT.EDU>
To: Greg Hudson <ghudson@mit.edu>
cc: athena10@mit.edu
In-Reply-To: <1207579052.5912.20.camel@error-messages.mit.edu>
Message-ID: <Pine.LNX.4.64L.0804071052340.9670@vinegar-pot.mit.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed
On Mon, 7 Apr 2008, Greg Hudson wrote:
> debathena-cluster-software is likely to be expanded to contain pretty
> much any native package someone requests be available on cluster
> machines. For example, http://diswww.mit.edu/menelaus/release-77/5904
> contains some requests from Alex Prengel.
>
> I don't want to couple this set to debathena-standard or
> debathena-workstation because people who want Athena-like machine
> behavior don't necessarily want multiple gigabytes of software just
> because someone wanted all of that stuff available on cluster machines.
>
> I also didn't want to tightly couple it to debathena-cluster (which
> doesn't exist yet, but is the name of the metapackage I expect to
> install on Debathena cluster machines), because someone might want to
> install this alongside debathena-workstation without turning their
> machine into a self-administering box. debathena-cluster will depend on
> debathena-cluster-software, of course.
Okay, debathena-cluster I guess was the missing piece in my understanding.
Seems like a reasonable plan.
>> debathena-standard depends debathena-clients depends debathena-hesiod-config depends hesiod
>
> Why does debathena-hesiod-config depend on hesiod? All hesiod contains
> is the "hesinfo" command, which is totally unnecessary for setting up
> the system Hesiod configuration.
Yeah, it should probably be moved to a direct descendent of
debathena-clients.
-Tim Abbott