It has been continuing to crash several times a day, and this time I had another machine on the same network easily accessible, and decided to poke around. X was eating the CPU, and wasn't responding to SIGTERM. SIGKILLing it worked, but gdm restarted it and it ate the CPU again. Killing it and then quickly stopping gdm stopped it starting again, and I was able to get a session on tty1 (albeit without any kind of display). Running textmode didn't get a display back, and neither did removing and inserting the i915 kernel module. So I ran /etc/acpi/sleep.sh, and woke it up again, and all is now good.
Unfortunately, the rapid Xorg restarts obliterate any useful debug logs of the initial crash from /var/log/Xorg*, but each hung start seems to leave the following lines near the end of the log, which aren't present otherwise: (WW) intel(0): PRB0_CTL (0x0001f001) indicates ring buffer enabled (WW) intel(0): PRB0_HEAD (0xee21abfc) and PRB0_TAIL (0x00000000) indicate ring buffer not flushed (WW) intel(0): Existing errors found in hardware state. -- -intel crashes regularly since Xserver 1.4 https://bugs.launchpad.net/bugs/162002 You received this bug notification because you are a member of Ubuntu Bugs, which is the bug contact for Ubuntu. -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
