Ferenc Wagner <wf...@niif.hu> writes: > So now I got another X server crash, which actually generated a core > file. Full backtrace below, it's lacking some frames, but maybe still > useful. Firefox didn't get any error report now, it simply hit the > closed connection together with the other desktop applications. The > problem didn't kill the X server instantly: after opening a new tab in > Firefox, severe screen corruption started to appear, during which I > could interact with the applications (at least the parts that weren't > corrupted beyond recognition). The server finally collapsed several > seconds later, then took considerable time to start again, and after > login it still isn't behaving normally: the cursor in urxvt is always > hollow (should be a filled block when focused) and off by one character > when backspacing, and selected text disappears instead of turning > inverse. Emacs and Firefox took enormous time to start, but are > perfectly OK and usable now (seems like they waited for something with > their windows half-drawn).
To be precise: in the restarted server there's still screen corruption on the second head (yes, I asked for the dual-head MGA driver) even now: applications work fine on the first head, but can't display on the second. Actually, urxvt manages pretty well (with the minor issues described above), Emacs, Firefox and Pidgin all fail miserably. They didn't take much time to start, as I stated above: they started all right, just couldn't display their final output. For some illustration (Pidgin on the right on the second head) see the Xinerama screenshot at http://apt.niif.hu/corrupt.png. -- Thanks, Feri. -- To UNSUBSCRIBE, email to debian-x-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87ipw1h5on....@tac.ki.iif.hu
