[CentOS-devel] CentOS-4.4 yumconf
cmc at math.hmc.edu
Tue Oct 17 07:46:31 UTC 2006
"CMC" == C.M. Connelly <cmc at math.hmc.edu>
"JH" == Johnny Hughes <mailing-lists at hughesjr.com>
I'm sorry if my comment was misconstrued -- I do appreciate the
(unpaid) time and energy you and others devote to CentOS, but as
an end user, while I can and do try to contribute in some ways
(such as filing bugs and reading and participating in mailing
lists), I cannot justify hanging out on IRC in the hope that
something I need to know might get discussed.
JH> If that is not the kind of OS you want, then you can pay
JH> $2500.00 a pop for a much more professional one. Oh, but
JH> every update cycle, they still make major changes. Look
JH> at the new things added this update cycle.
I did. I read through the upstream release notes before the
packages were built for CentOS. There's nothing in RH's notes
about the change in YUM/up2date configuration files.
Looking at the notes again now, I see that they have lots of
changelog extracts from various packages. But not for
redhat-release. The CentOS release notes also don't mention the
JH> That might be true and MANY things are worked out devel
JH> list. MANY, MANY things are discussed here. However, the
JH> yum configuration files, as well as the ones for up2date
JH> and any other update mechanism need to be in
JH> centos-release ... and that is where they are staying.
Ultimately I don't care where the files live, but it would have
been nice to have had a heads-up about the change so that my
machines didn't suddenly start updating from random mirrors. And
it would have been even nicer to have had some advice from the
CentOS developers on the ``right way'' to change the configuration
so that it wouldn't break.
JH> I apologize for not discussing this issue on this list
JH> before it was done, however it is still where those files
JH> belong. One of the things that is in our goals is to be
JH> more like upstream. It just makes sense to do it the same
JH> way where ever possible ... I am sorry that this affected
JH> some power users in an unexpected way, but it is still
JH> where the files belong.
As it happens, all it really did was to make my workstation and a
test machine start getting packages from other mirrors, as I
haven't yet rolled CentOS 4 out more widely.
I wasn't angry about the change, just a bit annoyed that I hadn't
known about it beforehand and that there didn't seem to be a clear
way to set things up so that they would work properly.
I think that we have that now (any of several possible solutions
that include removing CentOS-Base.repo and replace with an empty
file), so I'm satisfied. (Keeping the separate yumconf package
would have been easier for me, but I totally understand where
you're coming from in wanting to track upstream closely, and it's
clear from Fedora that that's where RH is going with these files.)
I'm sorry that you and Karanbir took my and others technical
criticism so personally. I *greatly* appreciate the work of the
CentOS developers -- you've saved me an enormous amount of work,
as the announcement of CentOS came when I was in the process of
doing a major reassessment of our OS plans after Red Hat had
announced the end of life of Red Hat Linux and before they'd said
anything about academic pricing. CentOS allowed me to build on
the work I'd already done to support deployment and maintenance
instead of starting from scratch with a new distro.
At this point, we could buy licenses from RH for a much lower
academic price, but the department doesn't really have that kind
of money. CentOS allows me to run an OS that has better support
than RH's current offerings for academia (i.e., *any*); is more up
to date than some other options (e.g., Debian), but still quite
stable; and also allows me to contribute something back to the
community in the form of feedback and bug reports, as well as the
mirror we provide for the releases and architectures that we
Thank you both for your work (as well as the other developers not
participating in this thread). And thank you for your apology for
not discussing the change on the list.
Claire Connelly cmc at math.hmc.edu
Systems Administrator (909) 621-8754
Department of Mathematics Harvey Mudd College
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 188 bytes
Desc: not available
More information about the CentOS-devel