[3121] in Release_7.7_team
Re: please create dev cell (beta) cluster information
daemon@ATHENA.MIT.EDU (Chris Terman)
Thu Jan 31 14:08:36 2002
Message-ID: <3C59944E.76541992@lcs.mit.edu>
Date: Thu, 31 Jan 2002 14:00:31 -0500
From: Chris Terman <cjt@lcs.mit.edu>
MIME-Version: 1.0
To: Mitchell E Berger <mitchb@mit.edu>
CC: cana@mit.edu, amb@mit.edu, wdc@mit.edu
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
ReSent-From: Cana Lynn McCoy <cana@MIT.EDU>
ReSent-To: <hesreq@mit.edu>, <release-77@mit.edu>
Mitchell E Berger wrote:
> Hope that clears things up... let us know if you've got other questions.
Thanks. I made the changes you suggested and seem
to be hooked into the right environment now.
We'll be reinitializing cs01 through cs40 with 9.0.24 over the
next few days. I've installed a cluster.local that gives us
local control of when updates happen, so I think we're pretty
much set to follow the Athena releases for the 6.004 machines.
- Chris
PS. We're still using our own kickstart installer which
works fine but it would be nice to transition to the IS
installer. Hopefully that'll work on our machine configurations
soon.
One issue: we do some minor customization of the lab machines,
so it would be convenient to be able to use our own rpm
list (essentially the usual release rpms + one our two of our
own). This would be doable with the current technology except
that it appears that the RPMS have to live somewhere in
the /afs/athena/system/rhlinux subtree. Is this right? Could
the mechanism be generalized to allow absolute pathnames
to RPMs (maybe it already does this?).
