> andrew wrote: > > On two different machines - one an old 32-bit AMD > the other a 64-bit Intel - I sometimes see X hang > after a user logs in. Is this a known problem with > build 130? > > There is a known bug in 130 that when Xorg crashes, > it hangs in > the cleanup/closedown of devices it does - look in > your /var/log/Xorg.0.log > to see if there's a backtrace showing it was > crashing. This is: > 6913965 Xorg 1.7.x goes into infinite loop in crash > h recovery > > That's fixed in 131, but that just makes it so the > server can exit (and > gdm can then start a new one) instead of hanging when > it segfaults. >
OK, this happened again on my M10 laptop. Xorg chews a CPU core with 100% user time. Only kill -9 gets rid of it, so I did that via SSH. X came back up fine, then when I logged out I got text mode screen with a flashing cursor. I then svcadm disabled gdm, then enabled and X came up, along with an "unable to authenticate user" error. I will try my 32-bit machine now, but I'm going to have to reinstall build 130 to do that unfortunately since I already deleted that BE. What info do you need from me to "root cause" this, and how do I get it? Thanks Andrew. > > Also, how do I get CTRL+ALT+BACKSPACE killing X > like it used to now XKB has taken over "dontzap" > duties? > > For now, with a HAL fdi file to enable the XKB > option. > > An example file is in: > http://mail.opensolaris.org/pipermail/desktop-discuss/ > 2009-August/013793.html > > but instead of 'ctrl:nocaps' you'd want > 'terminate:ctrl_alt_bksp' > > (Until Xorg 1.8, when input device configuration > moves back into xorg.conf > files with the impending death of HAL.) > -- > -Alan Coopersmith- > - alan.coopersmith at sun.com > Sun Microsystems, Inc. - X Window System > em Engineering > > _______________________________________________ > xwin-discuss mailing list > xwin-discuss at opensolaris.org -- This message posted from opensolaris.org
