Jaroslaw Polok wrote:

> Hi all
>
> Here's the proposal we've already started discussing
> during our Edinburgh HEPiX discussion on Scientific Linux
> about packaging additional kernel modules in uniform
> way:
>
> The most widely adopted format seems to be the one
> from Fedora Guidelines:
>
> http://www.fedora.us/wiki/PackagingKernelModules
>
> it allows for easy automating of kernel and associated
> kernel-module-foo package(s) using apt-get (checked ;-))
> and yum (read on a mailing list -> to be verified)
>
> Apparently 3rd party add-on package builders
> (http://dag.wieers.com/home-made/apt/mega-merge.php)
> also seem to agree (more-or-less) on this way -see
> repo-coord archives for
> May:http://lists.atrpms.net/pipermail/repo-coord/2004-May/
> (yes it is a flamewar running for couple of months now but calming
> down slowely : majority seems to support this option...)
>
> We at CERN currently packaged following in this format:
>
> kernel-module-bcm5700-kversion
> kernel-module-madwifi-kversion
> kernel-module-openafs-kversion
> kernel-module-xfs-kversion
> kernel-module-alsa-kversion
>
> This is quite nice as in order to install add-on module
> using apt I need to use only a command alike:
> apt-get install kernel-module-xfs-`uname -r` and the
> upgrade-kernel apt LUA script also handles upgrade
> of such packages along with kernel upgrade auto-magically.
>
> I think we could try to standardize on this packaging guideline
> for Scientific Linux too: what is your opinion ?
>
> (Most of these packages could go into SL contrib
> area as these are not really site specific -> but
> of course kernel dependent)
>
> Jarek [CERN Linux maintainer]
>
> __
> -------------------------------------------------------
> _ Jaroslaw_Polok ___________________ CERN - IT/ADC/LE _
> _ http://home.cern.ch/~jpolok ___ tel_+41_22_767_1834 _
> _____________________________________ +41_78_792_0795 _

I think this is a good idea.
I actually had been thinking of changing our AFS kernel modules to be
the same way CERN's is.  There just wan't time to get it into S.L. 3.0.1.
Troy

--
__________________________________________________
Troy Dawson  [log in to unmask]  (630)840-6468
Fermilab  ComputingDivision/CSS  CSI Group
__________________________________________________