[lug] Redhat 7.0

[Nate Duehr]

The changes to the LSB standards for the filesystem really screwed up a
machine at my desk at work.  Now partitions that didn't need much room
before, i.e. /var, now are the main "standard" place for things (as Debian's
been doing for a long time)... and the partitioning scheme is all wrong for
RH 7.0.  No warning, no courteous note: "Hey you!  We moved everything.  You
sure you want to upgrade, or would you rather read the readme at this URL
about it first?"

So that machine has mis-sized partitioning (with software RAID already set
up and doing its thing on what *USED* to be the important partitions now.
Good old RedHat'isms, /home/www and /home/ftp are in the "wrong" place
now... with no warning.  Of course I'm glad on the one hand that RH finally
got a clue and put them where everyone else agreed, but an UPGRADE should
mean just that... TELL me if you're going to point the ftp daemon to a
completely different place and replace my config files -- their presumption
that I want to go digging round for numerous .rpmsave files and fixing
things is a bad one.

That's just one of numerous things wrong with that system now, but probably
the most annoying.  I've found some things with file permissions that I
don't like, but I have to investigate how they got that way... was it the
upgrade, or does a clean load of RH 7 do that?

Samba, FTP, and other stuff pointed at /home/ftp/ broke immediately after
the upgrade and I had to waste my time fixing it instead of the guys at
RedHat waiting a few more weeks to release and doing some more scripting to
handle upgrades better.  They just didn't want to be shown up by Apple
releasing OS X in time for the RSA patent expiration.  There was NO need to
release nor any huge developments to suggest a roll from a 6.x number to 7.x
was necessary right now, other than the inclusion of OpenSSH.  Whooop deee
doo... OpenBSD has had it for a long time now.  It was the U.S. patent
system that had that one all messed up for us... it's not a reason for a
major revision number roll.

And releasing a version of GCC that the GCC team itself says isn't ready to
go out the door wasn't a good reason to roll that major number either.
IMHO.

I'm also glad the GCC guys went after them publically.  RedHat != Linux, and
they needed a reminder to communicate with the community writing the
software that made their company possible -- they don't set the standards,
the release schedules, or anything else except for the code they created and
added.  And even some of that was sent upstream to the various projects and
they don't have control of it anymore either.  :)

Nate

> From: Sean Reifschneider <jafo at tummy.com>
> Reply-To: lug at lug.boulder.co.us
> Date: Wed, 18 Oct 2000 18:47:48 -0600
> To: lug at lug.boulder.co.us
> Subject: Re: [lug] Redhat 7.0
> 
> On Wed, Oct 18, 2000 at 04:54:59PM -0700, Jeffrey B. Siegal wrote:
>> I would not upgrade to 7.0 until it stabilizes.
> 
> What issues have you seen with 7.0?
> 
> Sean
> -- 
> Rocky: "I must be getting near-sighted!  You look all fuzzy..."
> Bullwinkle: "Let's face it, Rock...  I *AM* all fuzzy."
> Sean Reifschneider, Inimitably Superfluous <jafo at tummy.com>
> tummy.com - Linux Consulting since 1995. Qmail, KRUD, Firewalls, Python
> 
> _______________________________________________
> Web Page:  http://lug.boulder.co.us
> Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
>