If I end a kde session from regular user, login as root, and then end session, x dies and you drop out to a console login. If I remove the synaptics driver from xorg.conf everything works fine (which is how I know the problem is with the synaptics driver). Reproducible: Always Steps to Reproduce: 1.End Session as regular user 2.Login as root user 3.End Session as root Actual Results: X dies and you end up back at the console login. Expected Results: You should have been brought back to the kdm login screen.
Created attachment 33267 [details] xorg log file
Created attachment 33268 [details] config file showing kernel configuration (2.6.5) to demonstrate the kernel config.
Created attachment 33269 [details] output from 'cat /proc/bus/input/devices'
Created attachment 33270 [details] output of dmesg
Created attachment 33271 [details] the wrong '/var/log/messages' file. Apologies.
Created attachment 33272 [details] the relevant xorg configuration
Pete, seen anything like this?
In fact, I uploaded the wrong messages (messages.out) file. I'll upload the correct one when I get home from work. Sorry for the trouble. J.K.
Created attachment 33404 [details] part of /var/log/messages relevant section; Jun 15 00:16:54 pantagruel kde(pam_unix)[6711]: session opened for user jkiely by (uid=0) Jun 15 00:17:00 pantagruel CRON[6837]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons ) Jun 15 00:17:30 pantagruel (jkiely-6939): starting (version 2.4.0.1), pid 6939 user 'jkiely' Jun 15 00:17:30 pantagruel (jkiely-6939): Resolved address "xml:readonly:/etc/gconf/gconf.xml.mandatory" to a rea$ Jun 15 00:17:30 pantagruel (jkiely-6939): Resolved address "xml:readwrite:/home/jkiely/.gconf" to a writable conf$ Jun 15 00:17:30 pantagruel (jkiely-6939): Resolved address "xml:readonly:/etc/gconf/gconf.xml.defaults" to a read$ Jun 15 00:18:00 pantagruel CRON[6948]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons ) Jun 15 00:18:16 pantagruel su(pam_unix)[6975]: session opened for user root by jkiely(uid=1000) Jun 15 00:19:00 pantagruel CRON[6995]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons ) Jun 15 00:20:00 pantagruel CRON[7007]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons ) Jun 15 00:21:00 pantagruel CRON[7018]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons ) Jun 15 00:22:00 pantagruel CRON[7031]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons ) Jun 15 00:22:15 pantagruel su(pam_unix)[6975]: session closed for user root Jun 15 00:22:33 pantagruel (jkiely-6939): Received signal 1, shutting down cleanly Jun 15 00:22:33 pantagruel (jkiely-6939): Exiting Jun 15 00:22:33 pantagruel kde(pam_unix)[6711]: session closed for user jkiely Jun 15 00:22:33 pantagruel mtrr: 0xe8000000,0x2000000 overlaps existing 0xe8000000,0x1000000 Jun 15 00:22:33 pantagruel mtrr: 0xe8000000,0x2000000 overlaps existing 0xe8000000,0x1000000 Jun 15 00:22:33 pantagruel agpgart: Found an AGP 2.0 compliant device at 0000:00:00.0. Jun 15 00:22:33 pantagruel agpgart: Putting AGP V2 device at 0000:00:00.0 into 4x mode Jun 15 00:22:33 pantagruel agpgart: Putting AGP V2 device at 0000:01:00.0 into 4x mode Jun 15 00:22:35 pantagruel kdm_greet[7247]: Can't open default user face Jun 15 00:22:41 pantagruel kde(pam_unix)[7246]: session opened for user root by (uid=0) Jun 15 00:23:00 pantagruel CRON[7346]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons ) Jun 15 00:24:00 pantagruel CRON[7359]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons ) Jun 15 00:24:29 pantagruel kde(pam_unix)[7246]: session closed for user root Jun 15 00:24:29 pantagruel kdm: :0[7381]: IO Error in XOpenDisplay Jun 15 00:24:29 pantagruel kdm[6707]: X server for display :0 terminated unexpectedly Jun 15 00:24:29 pantagruel kdm[6707]: Display :0 cannot be opened Jun 15 00:24:29 pantagruel kdm[6707]: Unable to fire up local display :0; disabling.
The ChangeLog from the authors site says, that an X-Server crash has been fixed with 0.13.4 http://w1.894.telia.com/~u89404340/touchpad/files/changes.txt How about this version?
This issue seems to have been fixed a while ago, and no additional info from the reporter for more than a year
