svlug-redhat_8.0_upgrade_problems_and_an_rpm_hint_about_zapping_stale_locks

This is part of The Pile, a partial archive of some open source mailing lists and newsgroups.




Date: Thu, 17 Oct 2002 18:23:42 -0700
From: "Daevid Vincent" <daevid@daevid.com>
To: "SVLUG" <svlug@lists.svlug.org>
Subject: [svlug] RedHat 8.0 continues to be problematic...

Well, I thought things were looking up. 

I downgraded Apache 2.0 back to the normal version to get PHP and mySQL
working. Learned a lot in the process about Apache ;-)

One thing I noticed is that Bluecurve graphical interface (while pretty)
doesn't have any icons at the bottom. All the 'spots' are just gears.
What happened there? Shouldn't there be something useful in each of
those spots?

KDE just doesn't work at all. Won't even load. :(

But here's the real kick in the balls. I noticed that my rhn account
showed 835 packages to be updated. Clearly this is an indication the
server and rhn are out of synch. So several times I try to have it
automate a synch. Each time I check back, I notice nothing has changed.

So I figure I'll remove the server from rhn and just re-register it from
scratch. No biggie right? I do the delete, then try to run
"rhn_register" and get a "command not found". Hmm... 

[root@daevid update]# locate rhn_register
/var/spool/up2date/rhn_register-gnome-2.7.9-7.x.2.i386.hdr
/var/spool/up2date/rhn_register-2.7.9-7.x.2.i386.hdr
/var/spool/up2date/rhn_register-2.7.9-7.x.2.i386.rpm
/var/spool/up2date/rhn_register-gnome-2.7.9-7.x.2.i386.rpm
/etc/sysconfig/rhn/rhn_register.rpmnew
/etc/sysconfig/rhn/rhn_register.rpmsave
/etc/pam.d/rhn_register.rpmnew
/etc/security/console.apps/rhn_register.rpmnew

Sure enough. There is no command? Okay, that sux... So I look on CD1 and
grab the grip of rhn_ files and figure I'll start to just 'rpm -Uvh'
them all again. Except the command is taking forever! I abort and try
some other .rpm files. Same thing. Any .rpm I try to install/update/ -qa
/ whatever just basically 'hangs'. I can't even "kill pid#" to get rid
of it.

Ugh. sigh.

Anyone know how I can get rpm's working again? Seems like a chicken/egg
situation. That is, how can I uninstall the /bin/rpm program and then
re-install from CD again, when it needs "itself" to install.

I'm scared to death to run the RH8.0 update (by inserting the CD into my
drive and turning the machine on) for fear that it will royally hose
everything it's taken so long to rebuild and get working (like it was
before the 'f**kgrade' from 7.3 to 8.0)

You'd think I'd learn my lesson not to trust an RHx.0 build, but I
figured after more than EIGHT versions, they would have their shit
together and the install would have gone more smoothly. Jeesh.

D.



--===============5965081979081539==
Content-Type: message/rfc822
MIME-Version: 1.0

Date: Thu, 17 Oct 2002 19:13:39 -0700
From: Karen Shaeffer <shaeffer@neuralscape.com>
To: svlug@lists.svlug.org
Subject: Re: [svlug] RedHat 8.0 continues to be problematic...
Message-ID: <20021017191339.A7168@synapse.neuralscape.com>
In-Reply-To: <001101c27644$ff5a5900$0500a8c0@telecom>;
	from daevid@daevid.com on Thu, Oct 17, 2002 at 06:23:42PM -0700
References: <001101c27644$ff5a5900$0500a8c0@telecom>
Content-Type: text/plain; charset=us-ascii
MIME-Version: 1.0
Precedence: list
Message: 2

On Thu, Oct 17, 2002 at 06:23:42PM -0700, Daevid Vincent wrote:
> 
> Anyone know how I can get rpm's working again? Seems like a chicken/egg
> situation.

As root:

# rm -f /var/lib/rpm/__db.*

This will eliminate the stale locks.

===

the rest of The Pile (a partial mailing list archive)

doom@kzsu.stanford.edu