abrt version: 1.1.14 architecture: i686 Attached file: backtrace cmdline: /usr/lib/openoffice.org3/program/soffice.bin -quickstart -nologo -nodefault comment: crash occured after a reboot. first time to happen. never happened before. component: openoffice.org crash_function: bridges::cpp_uno::shared::VtableFactory::initializeBlock executable: /usr/lib/openoffice.org3/program/soffice.bin kernel: 2.6.34.8-68.fc13.i686 package: openoffice.org-brand-1:3.2.0-12.35.fc13 rating: 4 reason: Process /usr/lib/openoffice.org3/program/soffice.bin was killed by signal 7 (SIGBUS) release: Fedora release 13 (Goddard) time: 1301019750 uid: 500 How to reproduce ----- 1. reboot system 2. ooo has crased msg 3.
Created attachment 487447 [details] File: backtrace
This looks like it originated in config. manager. The configmgr code has been rewritten for 3.3, because the old version was a nightmare, so it should be okay from F-14 up.
And maybe not... I forgot that we already had a few of these "initializeBlock" crashes in the past.
Do you have network-mounted home or any other unusual system configuration, by any chance?
*** Bug 575441 has been marked as a duplicate of this bug. ***
*** Bug 652725 has been marked as a duplicate of this bug. ***
Its a SIGBUS, I'd like to see what output on tail -f /var/log/messages (if any) appears between launch of OOo and final crash. So far the only ones of these that are reproducible was a disk throwing out i/o errors which fits with a SIGBUS.
(In reply to comment #4) > Do you have network-mounted home or any other unusual system configuration, by > any chance? Hello, No, I don't. I am on F-13 lxde on a different machine right now. I will run the command tomorrow on the F-14 lxde notebook. Caolon, do I just run the command $ tail -f /var/log/messages > /log.txt after logging in and before launching OOo?
tail -f /var/log/messages > /log.txt after logging in and before launching OOo yeah, just before launching OOo, e.g. tail -f /var/log/messages > /tmp/log.txt you may need to do the above as root.
I'm rather convinced these are generic hardware bugs, rather than OOo/LibreOffice ones. I'm almost tempted to stick a SIGBUS handled around initializeBlock and let it then fallback to trying elsewhere, but that way probably lies madness
*** Bug 710237 has been marked as a duplicate of this bug. ***
*** Bug 710241 has been marked as a duplicate of this bug. ***
*** This bug has been marked as a duplicate of bug 767708 ***