xfs - gui - recent updates

[Nigel Wade]

Gregory P. Ennis wrote:
> Dear List,
> 
> I have had two desktops that have had their gui stop working over the
> weekend.  Both were E-Machines with ATI Mach64 video cards using
> Princeton Ultra 72 and Ultra 74 monitors.  I have other E-Machines with
> different cards using the same monitors that are still working.
> 
> These two machines functioned without problems until the following
> updates occurred :  ( I can boot each to runlevel 3 without difficulty)
>  
> Jan 14 14:24:38 Updated: system-config-bind.noarch 4.0.0-38_FC4
> Jan 14 14:24:41 Updated: flex.i386 2.5.4a-35.fc4
> 
> The above only represents one machine, because I rebuilt the os on the
> other before I recognized it was not my mistake.  I currently do not
> have bind running nor do I use flex on either machine.  
>  
> Other observations that I have made include the X.org.0.log entry of:
> 
> Could not init font path element unix/:7100, removing from list!
> 
> Fatal server error:
> could not open default font 'fixed'
> 
> I have also noticed that the xfs daemon is not running and when I try to
> start it manually I get a "Segmentation Fault" error.
> 
> Has anyone else reported problems related to these updates.  Can any of
> you help me.  
> 
> Thanks,
> 
> Greg Ennis
> 
> 

I'm not running FC here at work, so can't give specific answers, but I might 
have some general pointers you could check.

When xfs starts it opens a unix domain socket, port 7100, to listen for requests 
(hence the "font path element unix/:7100" above). On RHEL this is in 
/tmp/.font-unix/fs7100. Does this file exist, is there a problem with the /tmp 
or /tmp/.font-unix directory? This is how /tmp/.font-unix and 
/tmp/.font-unix/fs7100 appear on my RHEL system:

drwxrwxrwt  2 root root 4096 Jan  4 08:54 /tmp/.font-unix/
srwxrwxrwx  1 root root 0 Jan 18 09:35 /tmp/.font-unix/fs7100

Can you run xfs manually, via strace, to see what system calls it makes, and if 
one is failing somewhere and causing it to abort? It might help pinpoint the 
location of the problem. A seg. fault is not normal when starting xfs ;-)

-- 
Nigel Wade, System Administrator, Space Plasma Physics Group,
             University of Leicester, Leicester, LE1 7RH, UK
E-mail :    nmw at ion.le.ac.uk
Phone :     +44 (0)116 2523548, Fax : +44 (0)116 2523555