Red Hat Bugzilla – Bug 197233
gconftool-2 hangs during 'yum update' under TWM
Last modified: 2008-05-01 11:38:06 EDT
Description of problem:
I'm running a freshly installed FC5 system. After several failed attempts
to get the screen-resolution app to behave as expected I found myself with
a Gnome environment that would not start after login. So I logged with
a Failsafe Terminal login and started TWM. Thinking that perhaps I should
pull the post-release updates to see if I could fix my resolution problems
I ran yum update through my squid proxy (using export http_proxy=http://...:3128).
Yum update chugged along happily until hitting the %post script for metacity
and hung with yum eating 100% of one of the CPUs. Bummer. I ran an rpm -q
--scripts on metacity and noticed the call to gconftool-2. Sure enough,
gconftool-2 was locked hard. Killing gconftool-2 caused yum to continue with
the next RPM.
I had similar lockups during the same session installing gnome-vfs2 and gnome-panel.
Two variables at play here:
1. I have a DISPLAY but GNOME is not running
2. I'm yumming through a proxy
None of the orb-o-matic stuff that gnome processes normally use to talk to
each other were running on the box at the time.
Same TWM login session. Same machine is dual boot with a newly installed FC4
on a different set of partitions. Under FC5 I mounted my
FC4 partitions, bind-mounted /proc and /sys and chrooted to my FC4 environment.
I was able to pull all of the past year's FC4 updates into my chroot without
encountering the same difficulties updating the packages that caused lockups
Version-Release number of selected component (if applicable):
GConf2-2.13.5-5 (maybe) during a global yum update that updated me
I happened this once... it's not the kind of thing I plan on doing often
Steps to Reproduce:
1. install fresh FC5/i386
2. log in as root under X with Failsafe Terminal and start TWM
3. set http_proxy to point at your local yumming squid
4. run yum update
RPMS that perform schema updates via gconftool-2 hang when attempting to do so
schema updates should complete without error
We no longer support Fedora Core 5 and I am currently trying to get my open bug
count down to a more manageable state. I'm going to close this bug as WONTFIX.
If this issue is still a concern for you, would you mind trying to reproduce on
a supported version of Fedora and reopening?
(this is a mass message)