Axel Thimm wrote:
> Hi,
> 
> On Tue, Jul 05, 2005 at 03:12:44PM -0500, Troy Dawson wrote:
> 
>>Axel Thimm wrote:
>>

..trimmed..

>>
>>Hi Axel,
>>It looks like you have everything as far as I can tell.  I don't think 
>>anyone could install from that tree ... but it certainly would make yum 
>>and apt transactions much easier.
> 
> 
> You mean an anaconda net-install? (I hope) that's covered with the
> mirror under http://dl.atrpms.net/mirrors/scientific/.
> 
> 
>>On a related note.  It looks like you are still moving things around, 
>>but where do you see the S.L. 3.x and 4.x compatible atrpms ending up 
>>at?  To me they look like they are at
>>
>>http://dl.atrpms.net/el3-i386/atrpms/
>>
>>and their comparable arch and distro places.  Is this where we should 
>>mirror?  Or do you see them changing to a different places such as
>>
>>http://dl.atrpms.net/sl3-i386/atrpms/
> 
> 
> Yes, you are right. The proper thing to do is to keep them under both
> (a hardlinked copy) to allow mirrorers to pick the desired distros
> easier instead of hunting them through the layout.
> 
> I'll fix this for atrpms and upcoming dag etc subdirs.
> 
> Does the layout otherwise make sense? It is still not in production in
> the sense that it hasn't been officially announced other than on
> atrpms-devel. I wouldn't really like to make any drastic changes
> shortly before going public though.
> 

Yes, I think the layout makes sense.
I think I'll just hold off on mirroring it until your done with all the 
changes.

> It cannot be used for anaconda net installs, because anaconda does
> have some strict requirements on the layout and only supports one
> repo. But if the plans/rumors on anaconda going yum turn out to become
> real, I hope that any yum repo will be anaconda plugable :)

That would be awsome and bizarre at the same time.  Because you 
potentially could install atrpm's, or dag's rpm's ... our the Fermi site 
rpm's, while doing a regular install.  The potential makes my little 
head swim. :)

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