J. K. Cliburn wrote:
I have a couple of old Dell Optiplex G110 PIII systems with Intel 810 onboard graphics that I use for testing purposes. I installed FC5 a couple of days ago, and now I've seen X crashes on both of them. Both are completely up to date with yum updates. The systems remain viable after the crash, that is, I can login with ssh from another host and poke around, but X is dead. Has anyone else seen this problem? Linux petrel 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux Here's the relevant syslog and Xlog snippets. First X crash host:gadwall /var/log/messages Mar 24 13:15:33 localhost gconfd (jcliburn-1927): Received signal 15, shutting down cleanly Mar 24 13:15:34 localhost gconfd (jcliburn-1927): Exiting Mar 24 13:15:41 localhost gdm[1849]: gdm_slave_xioerror_handler: Fatal X error - Restarting :0 Mar 24 13:15:48 localhost gdm[9796]: gdm_slave_xioerror_handler: Fatal X error - Restarting :0 Mar 24 13:15:54 localhost gdm[9830]: gdm_slave_xioerror_handler: Fatal X error - Restarting :0 Mar 24 13:15:54 localhost gdm[1819]: deal_with_x_crashes: Running the XKeepsCrashing script Mar 24 13:15:55 localhost gpm[1635]: *** info [client.c(137)]: Mar 24 13:15:55 localhost gpm[1635]: Connecting at fd 6 Second X crash host:gadwall Mar 25 06:44:03 localhost gconfd (jcliburn-12640): Received signal 15, shutting down cleanly Mar 25 06:44:03 localhost gconfd (jcliburn-12640): Exiting Mar 25 06:44:04 localhost gconfd (jcliburn-12848): starting (version 2.14.0), pid 12848 user 'jcliburn' Mar 25 06:44:04 localhost gconfd (jcliburn-12848): Resolved address "xml:readonly:/etc/gconf/gconf.xml.mandatory" to a read-only configuration source at position 0 Mar 25 06:44:04 localhost gconfd (jcliburn-12848): Resolved address "xml:readwrite:/home/jcliburn/.gconf" to a writable configuration source at position 1 Mar 25 06:44:04 localhost gconfd (jcliburn-12848): Resolved address "xml:readonly:/etc/gconf/gconf.xml.defaults" to a read-only configuration source at position 2 Mar 25 06:44:11 localhost gdm[1863]: gdm_slave_xioerror_handler: Fatal X error - Restarting :0 Mar 25 06:44:17 localhost gdm[12893]: gdm_slave_xioerror_handler: Fatal X error - Restarting :0 Mar 25 06:44:23 localhost gdm[12928]: gdm_slave_xioerror_handler: Fatal X error - Restarting :0 Mar 25 06:44:23 localhost gdm[1839]: deal_with_x_crashes: Running the XKeepsCrashing script Mar 25 06:44:24 localhost gpm[1656]: *** info [client.c(137)]: Mar 25 06:44:24 localhost gpm[1656]: Connecting at fd 6 Mar 25 06:44:35 localhost gconfd (jcliburn-12848): GConf server is not in use, shutting down. Mar 25 06:44:35 localhost gconfd (jcliburn-12848): Exiting /var/log/Xorg.0.log pgetbl_ctl: 0xb820001 pgetbl_err: 0x0 ipeir: 0 iphdr: 50c00004 LP ring tail: 538 head: 408 len: f001 start 3c0000 eir: 0 esr: 0 emr: 3d instdone: ff3a instpm: 0 memmode: 4 instps: 810 hwstam: 9ac7 ier: 0 imr: 9ac7 iir: 0 Fatal server error: Active ring not flushed Third X crash host:petrel /var/log/messages Mar 25 07:08:44 localhost gconfd (jcliburn-1798): Received signal 15, shutting down cleanly Mar 25 07:08:45 localhost gconfd (jcliburn-1798): Exiting Mar 25 07:09:20 localhost gconfd (jcliburn-2011): starting (version 2.14.0), pid 2011 user 'jcliburn' Mar 25 07:09:20 localhost gconfd (jcliburn-2011): Resolved address "xml:readonly:/etc/gconf/gconf.xml.mandatory" to a read-only configuration source at position 0 Mar 25 07:09:20 localhost gconfd (jcliburn-2011): Resolved address "xml:readwrite:/home/jcliburn/.gconf" to a writable configuration source at position 1 Mar 25 07:09:20 localhost gconfd (jcliburn-2011): Resolved address "xml:readonly:/etc/gconf/gconf.xml.defaults" to a read-only configuration source at position 2 Mar 25 07:09:27 localhost gconfd (jcliburn-2011): Resolved address "xml:readwrite:/home/jcliburn/.gconf" to a writable configuration source at position 0 Mar 25 07:38:20 localhost gconfd (jcliburn-2011): GConf server is not in use, shutting down. Mar 25 07:38:21 localhost gconfd (jcliburn-2011): Exiting /var/log/Xorg.0.log Error in I810WaitLpRing(), now is 832578981, start is 832576980 pgetbl_ctl: 0xb820001 pgetbl_err: 0x0 ipeir: 0 iphdr: 50c00004 LP ring tail: 7848 head: 76f0 len: f001 start 3c0000 eir: 0 esr: 0 emr: 3d instdone: ff3a instpm: 0 memmode: 4 instps: 810 hwstam: 9ac7 ier: 0 imr: 9ac7 iir: 0 space: 65184 wanted 65528 Fatal server error: lockup Error in I810WaitLpRing(), now is 832580990, start is 832578989 pgetbl_ctl: 0xb820001 pgetbl_err: 0x203000 ipeir: 0 iphdr: 50c00004 LP ring tail: 7850 head: 76f0 len: f001 start 3c0000 eir: 0 esr: 0 emr: 3d instdone: ff3a instpm: 0 memmode: 4 instps: 810 hwstam: 9ac7 ier: 0 imr: 9ac7 iir: 0 space: 65176 wanted 65528 FatalError re-entered, aborting lockup
I did notice a problem with the initial resolution being set to 800x600 after upgrading to FC5 from FC4. I ran system-config-display and selected the settings I prefer of 1280x1024 and maximum color resolution. The screen works fine now. I did not have a problem with crashing the server though.
Regarding the gconf error, all I found on a google search was related to an NFS mounted home directory.
If you saved your /home directory and someone else is logged in to the same account as in using vnc server or something, maybe the files are getting mangled in the ~/.gconf directory.
The 810 card might be lesser than the Intel 815 integrated video that my motherboard contains.
Did you try to move the ~/.gconf directory out of the way and launch X afterwards?
Anyway, the issue sounds more gnome related than X related in my novice opinion.
Jim -- Slight disorientation after prolonged system uptime is normal for new Linux users. Please do not adjust your browser. -- From a Slashdot.org post
