Hello list, hello redhat packagers (if any are reading this list),
<FLAME>
I work for a european e-broker where we have over 200 production
and dev servers with RH 6.1, 6.2, 7.1 and soon 7.2.. (we bought quite a
few rh boxed sets but no support, unfortunately). In order to support this
massive deployment and patch the servers, we have deployed a few install
servers which support installs and patches for client servers for multiple
arch's and RH versions.. That means that I'm following the patches for
every RH version very closely since we auto-modify the base distributions
on the install servers in order to provide pre-patched installable
distributions.. (except for RH6.2, of course, since the new rh6.2 packages
are built with rpm4 and this breaks the "instimage" directory on our
install servers so we had to freeze rh6.2 at 2.2.17-14 and have the new
installs patched afterwards)
I am annoyed at the apparent lack of serious thinking in the
numbering of the updates published by RedHat software for their supported
distributions.. Don't get me wrong, I'm a great fan of RedHat software and
I have always advocated RH where I had Linux servers to deploy over the
course of my small consulting carreer but could someone please explain the
numbering of some of the updated packages?
Here is what I mean (from RHSA-2001:106-08):
RH7.1 RH7.0 RH6.2
Prev. sendmail verion: 8.11.6-1.7.1 8.11.6-1.7.0 8.11.6-1.6.x
This sounds logical (1.7.1 sounds like version 1 of the package and it's
for rh7.1, 1.7.0 is version 1 of the package and it's for rh7.0, 1.6.x is
version 1 of the package and it's for all rh6.x versions - where x is 0,1
or 2).
Following this logic, it sounds logical that the latest sendmail update
from RH is numbered like this:
RH7.1 RH7.0
New sendmail verion: 8.11.6-2.7.1 8.11.6-2.7.0
But for some obscure logical reason, instead of calling the same update
"8.11.6-2.6.x" for RH6.x, it was decided by someone at RH to call the
update "8.11.6-1.6.y"? And even worse, in order to propagate the "error",
the same sendmail update for 5.x versions is also called ""8.11.6-1.5.y".
I didn't diff'ed the changelogs between the sendmail packages for all of
the dists but if the reason behind the numbering is that the fixes are
different, then these updates should not be part of the same RHSA
advisory..
I don't mind RH playing with the version numbers, it doesn't break "rpm
-Fvh *rpm", it doesn't break up2date (which I'm not using, mind you) it
doesn't break our own python scripts on our install servers but what is
the real logic behing this numbering? A commercial Linux distribution is
not a place where to have fun with numbers, that's what casinos are made
for.
</FLAME>
Thanks for reading.
Vincent, RHCE.
,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,
Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~
Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,.
Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'
http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ [EMAIL PROTECTED]
They cannot scare me with their empty spaces
Between stars - on stars where no human race is
I have it in me so much nearer home
To scare myself with my own desert places. - Robert Frost
_______________________________________________
Redhat-list mailing list
[EMAIL PROTECTED]
https://listman.redhat.com/mailman/listinfo/redhat-list
