27-nov-03
Hello Andre:
Thanks for the suggestion, but unfortunately it didn't work in my case. when I overwrote glibc-common, it errored on setting up links. I then tried overwriting glibc itself from the other machine and this rendered the machine unusable.
The problem may have been that the glibc version on the laptop I had used originally was glibc...i686 and the other machine had glibc...i386, as was glibc-common. This is probably what confused the link situation.
Anyway, I've reloaded the O.S. and will try the upgrade again, but in this case I'll do a
'rpm -Fvh glibc...i686 glibc-common...i386 glibc-devel...i386' and see how that goes.
Brgds
Bob
>From: Andre Costa
>Reply-To: rpm-list@xxxxxxxxxx
>To: rpm-list@xxxxxxxxxx
>Subject: Re: Rebuilding the database from scratch ??
>Date: Thu, 20 Nov 2003 09:58:52 -0200
>
>Hi Bob,
>
>yeah, that should fix it -- at least it worked in my case (thk God tar
>did not stop working). The steps I took were:
>
>on the 'healthy' machine:
>-----
>tar cvjf /tmp/glibc-ok.tbz `rpm -ql glibc`
>(repeat this step -- changing names accordingly -- to all glibc-*
>packages you have installed on the 'ill' machine)
>
>on the 'ill' machine:
>-----
>tar cvjf /tmp/glibc-bad-backup.tbz `rpm -ql glibc`
>(just in case... again, repeat to all glibc-* packages)
>cd /
>tar xvjf glibc-ok.tbz
>...
>(these came from the healthy machine)
>ldconfig -v
>(don't know if this is really necessary)
>
>There's something else you might want to try: if you have a RH-provided
>kernel on your bootloader configuration, you might want to try booting
>it, since it might work with the RH-only glibc.
>
>And, once your system is usable again, make sure you install the new
>(fixed) glibc RPMs, just so that RPM database stays consistent.
>
>HTH
>
>Andre
>
>On Thu, 20 Nov 2003 05:31:21 +0000
>"R.H. Palko" wrote:
>
> >
> > 20-nov-03
> >
> > Hello Andre:
> >
> > Thanks for the information, I believe you are right. I've found that
> > 'find' now hangs after finishing a search rather than exiting so its
> > affecting more than just rpm.
> >
> > I don't have the original glibc, but I do have a 'plain vanilla' RH 9
> > box which was successfully updated to glibc* 2.3.2-27.9.7 so I'll tar
> > up the glibc-common and glibc-devel from that box and I'll try
> > overwriting the laptop with those which should bring everything up to
> > 2.3.2-27.9.7 and hopefully fix the problem.
> >
> > Brgds
> > Bob
> >
> >
> > >From: Andre Costa
> > >Reply-To: rpm-list@xxxxxxxxxx
> > >To: rpm-list@xxxxxxxxxx
> > >Subject: Re: Rebuilding the database from scratch ??
> > >Date: Tue, 18 Nov 2003 09:20:47 -0200
> > >
> > >Hi folks,
> > >
> > >thought it might help: I had the exact same probls when I upgraded my
> > >glibc from RH a couple of days ago -- it happened that 2.3.2-27.9.6
> > >was only compatible with RH-kernels, and my vanilla 2.4.22 did not
> > >work with that. As a result, _many_ apps (rpm, gkrellm, opera etc.)
> > >started segfaulting -- rpm segfaulted at the very same point you
> > >describe below, leaving behind a __db.001 file.
> > >
> > >Fortunately I had access to another RH9 box which hasn't been
> > >updated, and I made a tarball with all the files listed by 'rpm -ql
> > >glibc', and installed it over the files at my box. I don't know if
> > >would have been able to recover otherwise...
> > >
> > >RH released a new version (2.3.2-27.9.7) one day after, for which the
> > >only difference was:
> > >
> > >[Update 2003-11-13]: The packages for Red Hat Linux 9 have been
> > >updated for compatibility with kernels not provided by Red Hat.
> > >
> > >I installed it on my box and everything was smooth again.
> > >
> > >BTW: I am now running Fedora.
> > >
> > >HTH
> > >
> > >Andre
>
>--
>Andre Oliveira da Costa
>
>
>_______________________________________________
>Rpm-list mailing list
>Rpm-list@xxxxxxxxxx
>https://www.redhat.com/mailman/listinfo/rpm-list
Protect your PC -
Click here for McAfee.com VirusScan Online
_______________________________________________
Rpm-list mailing list
Rpm-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/rpm-list
