[7251] in testers
Re: rpmdb corruption
daemon@ATHENA.MIT.EDU (Jonathon Weiss)
Tue Jul 12 20:49:01 2005
Message-Id: <200507130048.j6D0mnfg014928@the-other-woman.mit.edu>
From: Jonathon Weiss <jweiss@MIT.EDU>
To: Jonathon Weiss <jweiss@MIT.EDU>
cc: testers@MIT.EDU
In-reply-to: Your message of "Sat, 09 Jul 2005 07:58:32 EDT."
<200507091158.j69BwWGt003734@bearing-an-hourglass.mit.edu>
Date: Tue, 12 Jul 2005 20:48:49 -0400
>
> I tried to take 9.4.10 on vorpal-blade, and the update failed because
> of a fissing findutils package. Note that I've never un-installed
> this package, and all of the files that are in it exist on vorp.
>
> /etc/athena/rpmupdate -n -p /dev/null /var/athena/release-rpms
>
> confirmed that findutils was the only missing package, but removing
> the -n caused it to spew errors of the form:
>
> error: rpmdbNextIterator: skipping h# 2770 Header V3 DSA signature: BAD, key ID db42a60e
>
> running 'rpm --rebuilddb' allowed rpmupdate to re-add the findutils
> rpm, and allowed me to proceed to the next error:
>
> vorpal-blade.mit.edu# update_ws
> Beginning update from 9.4.9 to 9.4.10 at Sat Jul 9 07:25:56 EDT 2005.
> Version: $Id: update,v 1.20 2005/07/08 17:12:56 ghudson Exp $
> warning: rhel-4-updates/RPMS/cpp-3.4.3-22.1.i386.rpm: V3 DSA signature: NOKEY, key ID db42a60e
> error: rhel-4-updates/RPMS/glibc-common-2.3.4-2.9.i386.rpm: V3 DSA signature: BAD, key ID db42a60e
> rpmupdate: Invalid package file rhel-4-updates/RPMS/glibc-common-2.3.4-2.9.i386.rpm
> *** The update has failed ***
> Please contact Athena Cluster Services at x3-1410. -Athena Operations
> Update complete; please reboot for changes to take effect.
>
> Whcih after some fumbling I fixed by stoping afs, mkfs-ing the cache
> partition and rebooting.
>
> Unfortunately, my machine then hung during the update (it hung last
> weekend too, and I currentyl suspect a hardware problem, but that's a
> guess. I guess I will be sad now.
>
> Jonathon
>
>
OK, at this point I'm fairly convinced that I have a hardware problem
that's causing data read from the disk to be corrupted by the time the
CPU is done with it. All of the above should be disregarded.
Jonathon
