Yeah the problem is that the dump tool needs to allocate memory to create the dump info but sometimes the system is in such a state that it can't do it, so the tool obviously fails.
But for GPU lockup bugs with Intel graphics, all you really need to collect is the output of 'dmesg' and your /sys/kernel/debug/dri/0/i915_error_state file. Both of these must be collected while the machine is locked up (e.g. by sshing into the sick machine over ethernet), but that should work even if the dump tool is failing. See https://wiki.ubuntu.com/X/Troubleshooting/Freeze for additional info. If you do that, please file a NEW bug report for the freeze. We'll focus this bug report on the issue that apport-gpu-error-intel.py fails under these conditions. I need to rejigger the script to stop using the dump tool and instead use the error file directly. ** Package changed: xserver-xorg-video-intel (Ubuntu) => xdiagnose (Ubuntu) ** Changed in: xdiagnose (Ubuntu) Importance: Undecided => High ** Changed in: xdiagnose (Ubuntu) Status: New => Triaged ** Changed in: xdiagnose (Ubuntu) Milestone: None => oneiric-alpha-1 ** Changed in: xdiagnose (Ubuntu) Assignee: (unassigned) => Bryce Harrington (bryce) -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/776580 Title: apport-gpu-error-intel.py has never succeedded on gpu error on this computer -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
